JPET

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Brambilla, L.
Right arrow Articles by Cantoni, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Brambilla, L.
Right arrow Articles by Cantoni, O.

Vol. 284, Issue 3, 1112-1121, March 1998

Electron Transport-Mediated Wasteful Consumption of NADH Promotes the Lethal Response of U937 Cells to Tert-Butylhydroperoxide

Liliana Brambilla, Piero Sestili, Andrea Guidarelli, Letizia Palomba and Orazio Cantoni

Istituto di Farmacologia e Farmacognosia and Centro di Farmacologia Oncologica Sperimentale, Università di Urbino, Urbino, Italy


    Abstract
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
Conclusions
References

The toxicity of a short-term exposure to tert-butylhydroperoxide in U937 cells was markedly reduced by chemically or experimentally induced respiratory deficiency. Rotenone mitigated the lethal effects of the hydroperoxide over the same concentration-range in which the complex I inhibitor inhibited oxygen utilization. U937 cells that were made respiration deficient by growing them in the presence of either chloramphenicol or ethidium bromide, were in both circumstances highly resistant to tert-butylhydroperoxide. The improved survival was not a direct consequence of the absence of electron transport, but rather was attributable to the large amounts of NADH which accumulate in the mitochondria of chemically hypoxic or respiration-deficient cells. Indeed, the toxicity elicited by tert-butylhydroperoxide was also abolished by supplementation with either of two different NADH-linked substrates, namely pyruvate or beta -hydroxybutyrate. Accumulation of intramitochondrial NADH, and the resulting cytoprotective effects, was associated with prevention of the loss of nonprotein sulphydryls and mitochondrial membrane potential. Neither rotenone nor pyruvate reduced the toxicity of tert-butylhydroperoxide in thiol-depleted cells. Taken together, these results indicate that depletion of mitochondrial NADH is a critical event in the cytotoxic response to tert-butylhydroperoxide since this pyridine nucleotide prevents mitochondrial dysfunction and cell death caused by the hydroperoxide. As a consequence, in hydroperoxide-treated cells electron transport is highly detrimental since it consumes mitochondrial NADH.


    Introduction
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
Conclusions
References

The organic hydroperoxide tB-OOH induces an array of cellular dysfunctions including peroxidation of membrane lipids (Masaki et al., 1989a), depletion of GSH (Geiger et al., 1993; Korytowski et al., 1995), oxidation of NAD(P)H (Bellomo et al., 1984; Livingston et al., 1992), perturbation of calcium ion sequestration (Livingston et al., 1992), DNA single strand breakage (Sandström and Marklund, 1990; Sandström, 1991; Baker and He, 1991; Guidarelli et al., 1995) and mitochondrial damage (Masaki et al., 1989b; Wu et al., 1990; Imberti et al., 1993; Castilho et al., 1995; Nieminen et al., 1995). This wide spectrum of effects well emphasizes the high reactivities of the tB-OOH-derived radical species and may represent a valid argument supporting the notion that all, or at least some, of the above effects may play a role in the cytotoxic response to the hydroperoxide. Some of the deleterious effects of tB-OOH, including DNA single strand breakage, have however recently been shown to be ineffectual in terms of cell killing. Indeed, although considered by some authors (Baker and He, 1991) a potentially toxic event, this type of lesion was found to be insensitive to antioxidants under the same experimental conditions in which the lethal response was prevented (Coleman et al., 1989; Guidarelli et al., 1995). Recent work from our laboratory (Guidarelli et al., 1996a) has confirmed these findings by showing that the complex III inhibitor antimycin A reduced the formation of tB-OOH-derived radical species as well as the cytotoxic response, but increased DNA strand scission. Thus, different species appear to mediate DNA cleavage and cytotoxicity in cells exposed to tB-OOH. Although the species producing DNA damage are still a matter of debate and currently under active investigation, it is generally agreed (see above) that the species producing cell death are represented by the tert-butoxyl and methyl radicals (Iannone et al., 1993; Barr and Mason, 1995; O'Donnell and Burkitt, 1994; Guidarelli et al., 1996a). It has also been suggested (O'Donnell and Burkitt, 1994; Guidarelli et al., 1996a) that at least a proportion of these species are generated within the mitochondria.

In recent years the effects of tB-OOH on mitochondrial structure and functions have been extensively investigated and these studies have demonstrated that the hydroperoxide impairs mitochondrial ATP synthesis (Imberti et al., 1993). This effect is expected to be detrimental, because these conditions prevent repair processes as well as the execution of the functions responsible for maintaining cell viability. Indeed, the lethal response of cultured hepatocytes exposed to tB-OOH was markedly reduced by the glycolytic substrate fructose (Imberti et al., 1993) and, as a consequence, depletion of ATP was suggested to play a pivotal role in the mechanism whereby tB-OOH induces cell killing (Nieminen et al., 1990; Imberti et al., 1993; Toussaint et al., 1994; Nieminen et al., 1995). High concentrations of tB-OOH also lead to mitochondrial permeability transition and this effect may well represent a cause of cell death since cyclosporin A, an agent which binds to cyclophillin thereby preventing pore opening (Halestrap and Davidson, 1990), inhibited both responses (Nieminen et al., 1995).

In summary, the current knowledge in the literature is consistent with a model in which mitochondria play a central role in the mechanism of cell death promoted by tB-OOH. According to this model the hydroperoxide-derived cytotoxic intermediates are generated within the mitochondria and cause mitochondrial dysfunction and finally opening of permeability transition pores. As a consequence, a number of critical reactions take place in the mitochondria of tB-OOH-intoxicated cells and the lethal response can be expected to result from the balance between events in which tB-OOH-derived radical species are being generated and produce deleterious effects and events involved in the removal of these lesions.

This premise represents the basis of our work in which we investigated the role of mitochondrial NADH in the cytotoxic response to tB-OOH. We tested the hypothesis that the mitochondrial fraction of this pyridine nucleotide serves as a vital source of reducing equivalents counteracting the sequence of events leading to mitochondrial dysfunction and to cell death upon exposure to high levels of tB-OOH.

    Materials and Methods
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
Conclusions
References

Materials. Pyruvate, beta -OHB, tB-OOH, H2O2, BSO, rotenone and the remaining chemicals were from Sigma-Aldrich, Milan, Italy. RPMI 1640 culture medium was from GIBCO, (Grand Island, NY) and fetal bovine serum, penicillin and streptomycin were from Serelab (Sussex, UK). T-75 tissue culture flasks were purchased from Corning (Corning, NY). [14C]-thymidine was obtained from NEN/Du Pont (Boston, MA). Y.S.I. oxygraph (model 5300) and Supelcosil LC-18 column were purchased from Yellow Springs Instruments Co. (Yellow Springs, OH) and Supelco (Bellafonte, PA), respectively. Laser confocal microscope (View Scan DVC-250) was obtained from Bio-Rad (Richmond, CA).

Cell culture and treatments. Human myeloid leukemia U937 cells were grown in RPMI 1640 culture medium supplemented with 10% (v/v) fetal bovine serum, penicillin (50 U/ml), and streptomycin (50 µg/ml), at 37°C in T-75 tissue culture flasks in an atmosphere of 95% air-5% CO2.

Respiration-deficient U937 cells were isolated by culturing the cells in RPMI medium containing 110 µg/ml pyruvate, 5 µg/ml uridine and 400 ng/ml ethidium bromide (EB cells) or 50 µg/ml chloramphenicol (CP cells) for 6 days with medium changes every 2 days. When respiration-proficient and -deficient cells are discussed together, they are indicated in the text as RP, EB and CP cells, respectively; where respiration-proficient cells alone are discussed, they are indicated as U937 cells.

In those experiments involving GSH depletion, U937 cells (5 × 106/20 ml) were incubated for 16 hr at 37°C in RPMI medium containing 10 µM of BSO, an inhibitor of gamma -glutamylcysteine synthetase (Griffith and Meister, 1979).

Stock solutions of tB-OOH, H2O2, pyruvate and beta -OHB were freshly prepared in Saline A (0.14 M NaCl, 5 mM KCl, 4 mM NaHCO3); rotenone was dissolved in 95% ethanol. At the treatment stage the final ethanol concentration was never higher than 0.05%. Under these conditions ethanol was neither toxic nor DNA-damaging, nor did it affect the cyto-genotoxic properties of tB-OOH.

Treatments were performed in Saline A both in the absence or presence of glucose (5 mM). At the treatment stage, the cell density was of 2 × 105 cells/ml. Cells were used at a density of 1 × 105/ml only in those experiments involving exposure to H2O2.

Cytotoxicity assay. After the treatments, the cells were washed with ice-cold glucose-containing or -free Saline A and resuspended in prewarmed RPMI 1640 culture medium, plated into 35-mm tissue culture dishes and incubated at 37°C for 6 hr. Cytotoxicity was determined using the trypan blue exclusion assay. Briefly, an aliquot of the cell suspension was diluted 1:1 (v/v) with 0.4% trypan blue and the cells were counted using a hemocytometer. Results are expressed as the percentage of dead cells (ratio of stained cells vs. the total number of cells).

Oxygen consumption. Cells were washed once in glucose-free Saline A and suspended in the same buffer at a density of 1 × 107 cells/ml. Oxygen consumption was measured using a Y.S.I. oxygraph equipped with a Clark-type electrode at 37°C. The cell suspension (3 ml) was transferred to the polarographic cell and, under constant stirring, the basal respiration was measured immediately after addition of 5 mM glucose over a 3-min time interval. After basal respiration, the NADH-linked substrates or rotenone were added and the rate of oxygen utilization was monitored for a further 3 min and calculated as described by Robinson and Cooper (1970).

NPSH assay. NPSH levels were determined using the 5, 5'-dithiobis-(2-nitrobenzoic acid) method, as previously described (Cantoni et al., 1994). The concentration of NPSH in the samples was determined by comparison with standard solutions of GSH. Protein contents were assayed as described by Lowry et al. (1951) using bovine serum albumin as standard.

Determination of adenine nucleotides. NADH and ATP levels were assayed as described in (Stocchi et al., 1985) using alkaline or perchloric acid extraction, respectively. The aqueous extracts were analyzed for their NADH and ATP contents by reversed-phase high performance liquid chromatography using a Supelcosil LC-18 column. The proteins retained in the Centricon-50 Amicon filter and those contained in the perchloric acid precipitate were determined using the Lowry assay (Lowry et al., 1951).

Alkaline elution assay. Cells were labeled overnight with [14C]thymidine (0.05 µCi/ml) and incubated for a further 6 hr in a medium containing unlabelled thymidine (1 µg/ml) before treatments and subsequent analysis for DNA damage. The filter elution assay was carried out by a procedure virtually identical to that described in (Kohn et al., 1981) with minor modifications (Cantoni et al., 1986). Strand Scission Factor values were calculated from the resulting elution profiles by determining the absolute log. of the ratio of the percentage of DNA retained in the filters of the drug-treated sample to that retained in the untreated control sample (both after 8 hr of elution).

tert-Butylhydroperoxide assay. tB-OOH levels were assayed spectrophotometrically according to the method of Heath and Tappel (1976). Briefly, the cells were treated with 1 mM tB-OOH in the presence or absence of 0.5 µM rotenone or 5 mM pyruvate in glucose-containing Saline A at 37°C. At the indicated times the cells were rapidly collected by centrifugation (1200 rpm for 5 min) and a 50-µl aliquot of the supernatant was diluted to 1 ml with a solution (pH 7.5) of a 0.1 M phosphate buffer containing 2 mM EDTA, 160 µM NADPH, 2 mM GSH and 1 U/ml glutathione reductase. Samples were equilibrated for 2 min and the reaction was started by adding 5 mU GPx. Absorbance of residual NADPH was then monitored after 5 min at 340 nm.

GPx activity assay. Cell pellets obtained after centrifugation at 1200 rpm for 5 min were washed twice with ice-cold phosphate buffered saline (0.121 M NaCl, 10 mM NaH2PO4, 1.5 mM KH2PO4, 3 mM KCl), resuspended in distilled water at a final density of 2 × 106 cells/100 µl and sonicated three times for 15 sec at 20 W (using a Branson sonifier). GPx activity was estimated by the method of Gunzler et al. (1974), which involves monitoring at 340 nm of the disappearance of NADPH in a reaction scheme that couples the GPx-mediated GSH-dependent reduction of tB-OOH and the glutathione reductase-mediated NADPH-dependent reduction of oxidized glutathione.

Mitochondrial membrane potential. U937 cells (2 × 105/ml) were exposed for 5 min to 5 mM pyruvate or 0.5 µM rotenone or for 15 min to 0.5 µM cyclosporin A and then treated with 1 mM tB-OOH for 30 min in glucose-containing Saline A at 37°C. The cells were then washed and resuspended in prewarmed RPMI 1640 culture medium, plated into 35-mm dishes and incubated at 37°C for 4.5 hr. Rhodamine 123 (10 µg ml) was added to the cultures 30 min before the end of the postincubation. The cells were then washed twice with phosphate buffered saline and resuspended in 100 µl of the same buffer; 20 µl (80,000 cells) of this cell suspension was stratified on a slide and excited with an Argon laser. The analysis was performed with a laser confocal microscope. Mitochondrial retention of rhodamine 123 was taken as an index of mitochondrial polarization (Lemasters et al., 1987).

Statistical analysis. Statistical analysis of the data for multiple comparisons was performed by analysis of variance followed by Dunnett' s test. For single comparison, the significance of differences between means was determined by Student's t test.

    Results
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
Conclusions
References

The complex I inhibitor rotenone prevents cell death induced by tB-OOH under the same conditions in which it abolishes oxygen consumption. Rotenone is a potent inhibitor of the mitochondrial respiratory chain and has the great advantages of being both membrane-permeant and a selective poison for complex I. Indeed, the addition of rotenone to the respiratory medium (glucose-containing Saline A) promoted a prompt and concentration-dependent decrease in U937 cell oxygen consumption (fig. 1A). Obviously, in the absence of glucose the rate of oxygen utilization was considerably (about three times) lower and was also sensitive to inhibition by rotenone (not shown). Having established the conditions resulting in inhibition of oxygen consumption, we then investigated the role of electron transport in the toxic response of U937 cells to challenge with tB-OOH. For this purpose, the cells were exposed for 30 min to increasing concentrations of the peroxide under conditions of low basal oxygen consumption, i.e., in the absence of glucose, and under more physiological conditions involving the presence of glucose (5 mM). The number of dead cells was estimated after 6 hr of posttreatment incubation in complete culture medium. The results illustrated in figure 1B clearly indicate that the toxic response to tB-OOH increased as a function of the peroxide concentration and was always more pronounced when treatments were performed in the absence of glucose. Addition of 0.5 µM rotenone abolished the cytotoxicity promoted by tB-OOH both in the absence and presence of glucose. The inset to figure 1B indicates that cell killing promoted by H2O2 (1 mM) was both glucose-independent and not affected by rotenone.


View larger version (21K):
[in this window]
[in a new window]
  		Fig. 1.   Rotenone concentration-dependence for inhibition of U937 cell oxygen consumption and tB-OOH-induced toxicity. A, U937 cell oxygen consumption was assessed as detailed in "Materials and Methods." Basal respiration (12.7 ± 0.39 nmol O2/min/107 cells) was recorded for 3 min in the presence of 5 mM glucose. The effect of increasing concentrations of rotenone was then measured. Experimental results represent the mean of two separate experiments. B, U937 cells (2 × 105/ml) were exposed for 30 min to increasing concentrations of tB-OOH in the absence (open symbols) or presence (closed symbols) of 0.5 µM rotenone. Treatments were performed in Saline A with (circles) or without (squares) glucose (5 mM) and were followed by a 6-hr postincubation in drug-free culture medium. Toxicity was then assessed using the trypan blue exclusion assay. Rotenone, at all the concentrations utilized in these experiments, did not produce cell killing. The inset shows the results of similar experiments in which treatments were performed with 1 mM H2O2 in the absence (open bars) or presence (solid bars) of 0.5 µM rotenone using a cellular density of 1 × 105 cells/ml. Results represent the mean ± S.E.M. calculated from five to seven separate experiments, each performed in duplicate. The toxicity of tB-OOH was significantly reduced by glucose at *P < .0005 (unpaired t test). The toxicity elicited by all the concentrations of tB-OOH (both in the absence and presence of glucose) was significantly reduced by rotenone at *P < .0005 (unpaired t test). C, Cells were exposed to 1 mM tB-OOH in the presence of increasing concentrations of rotenone. Treatments were performed as described in B with (closed circles) or without (closed squares) 5 mM glucose. After treatment, the cells were rinsed and postincubated for 6 hr in complete culture medium. The number of dead cells was then assessed by the trypan blue exclusion method. Results represent the mean ± S.E.M. calculated from three to seven separate experiments, each performed in duplicate. The inset shows a correlation plot relating the effects of increasing concentrations of rotenone on oxygen consumption (data from A) and on the toxicity of tB-OOH (data from C). The toxicity of tB-OOH (given with or without glucose) was significantly reduced by rotenone at *P < .01 (analysis of variance followed by Dunnett's test).

We next investigated the rotenone concentration-dependence for inhibition of cell death caused by tB-OOH (1 mM) and found that cytoprotection was apparent using levels of the inhibitor as low as .025 µM and was maximal at .5 µM (fig. 1C). It is interesting to observe that rotenone inhibited tB-OOH cytotoxicity over the same concentration-range in which inhibition of oxygen consumption was detected (fig. 1A). Indeed, as illustrated in the inset to figure 1C, a linear relationship (r = 0.96) exists between these two parameters. The results reported in figure 1C also indicate that rotenone abrogated the glucose-dependence of the cytotoxic response to tB-OOH.

Respiration-deficient U937 cells are remarkably resistant to killing promoted by tB-OOH

As an alternative approach for the assessment of the role of electron transport in the toxic response of U937 cells to challenge with tB-OOH, we measured the killing induced by the hydroperoxide in respiration-deficient cells. U937 cells were made respiration-deficient using the two different approaches described in "Materials and Methods." We found that oxygen consumption stimulated by glucose, either alone or associated with pyruvate, was below the detection limits after growth in RPMI medium containing 110 µg/ml pyruvate, 5 µg/ml uridine and 400 ng/ml ethidium bromide (EB cells) or 50 µg/ml chloramphenicol (CP cells) for as little as 4 days (not shown).

The results reported in figure 2 clearly demonstrate that experimentally induced respiratory deficiency protects against cell death induced by tB-OOH although not affecting the toxicity generated by H2O2 (inset). Figure 2 also shows the lethal response of respiration-proficient U937 cells treated with concentrations of tB-OOH higher that 1 mM in the presence of .5 µM rotenone. It is intriguing to observe that the toxicity curves obtained with EB and CP cells exposed to tB-OOH, or respiration-proficient cells treated with the hydroperoxide and rotenone, are basically superimposable.


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 2.   Experimentally induced respiratory deficiency confers resistance to the lethal effect of tB-OOH. RP (circles) or EB (triangles) cells (2 × 105/ml) were exposed for 30 min in glucose-containing Saline A to increasing concentrations of tB-OOH in the absence (open symbols) or presence (closed symbols) of 0.5 µM rotenone and then grown for further 6 hr in fresh culture medium. The lethal effect of tB-OOH on CP cells was also investigated (squares). The inset shows the cytotoxic response of RP (circles) and EB (triangles) cells (1 × 105/ml) to increasing concentrations of H2O2. Cytotoxicity was determined as detailed in the legend to figure 1 (B). Each experimental point detailed in the main graph represents the mean ± S.E.M of three to five separate determinations, each performed in duplicate. Experimental results shown in the inset are the mean of two separate determinations, each performed in duplicate. *P < .01 compared with RP cells treated with tB-OOH alone by analysis of variance followed by Dunnett's test.

Finally, we investigated the effect of rotenone on the toxicity of tB-OOH in EB cells and found that, under these conditions, the complex I inhibitor was unable to prevent the cell killing promoted by the hydroperoxide (fig. 2).

Rotenone or experimentally induced respiratory deficiency does not mitigate the induction phase of the deleterious effects promoted by tB-OOH. The experiments reported in this section were designed to assess whether suppression of electron transport directly or indirectly modifies the extent of the insult received by the cells upon exposure to tB-OOH. The results illustrated in figure 3A demonstrate that respiration-proficient and -deficient U937 cells were equally sensitive to DNA single-strand breakage triggered by tB-OOH and that this event was not affected by rotenone.


View larger version (21K):
[in this window]
[in a new window]
  		Fig. 3.   Rotenone as well as experimentally induced respiratory deficiency affects neither tB-OOH metabolism nor the formation of DNA lesions or inhibition of GPx activity mediated by the hydroperoxide. A, RP (circles) or EB (triangles) cells were challenged with increasing concentrations of tB-OOH for 30 min in glucose-containing Saline A in the absence (open symbols) or presence (closed symbols) of 0.5 µM rotenone, and then analyzed for DNA single strand breakage with the alkaline elution technique. Results represent the mean ± S.E.M. of three separate determinations, each performed in duplicate. B, RP (open circles) or EB (open triangles) cells, were exposed to 1 mM tB-OOH in glucose-containing Saline A, at a density of 2 × 105 cells/ml. Initial and residual concentrations of tB-OOH were then measured at the indicated time points as detailed in "Materials and Methods." The effect of rotenone (0.5 µM) on the metabolism of tB-OOH in RP cells was also investigated (closed circles). Results represent the mean ± S.E.M. of three separate determinations, each performed in duplicate. The inset shows the extent of GPx-inhibition induced by tB-OOH (1 mM for 15 min) in RP cells in the absence or presence of rotenone (0.5 µM). Results are expressed as percent GPx activity of sham-treated cells (114.7 mU/106 cells) and represent the mean values from two separate experiments with similar outcomes.

The rate of disappearance of tB-OOH from the culture medium of respiration-proficient and -deficient U937 cells was also investigated. For this purpose, the cells were treated with 1 mM tB-OOH in a glucose-containing saline, and the levels of the hydroperoxide in the culture medium were measured immediately after addition of the drug as well as after various time intervals. The same experiment was repeated using respiration-proficient cells exposed to tB-OOH in the presence of 0.5 µM rotenone. As illustrated in figure 3B neither experimentally induced respiratory deficiency nor the complex I inhibitor significantly altered the rate of disappearance of tB-OOH from the culture medium, strongly suggesting a lack of relationship between cytoprotection and effects on hydroperoxide metabolism. In figure 3B it can also be seen that tB-OOH was efficiently metabolized within the first 10 min of treatment whereas at later times peroxide removal was virtually stopped. These results are readily explained by the observation that tB-OOH inhibits GPx activity (inset to fig. 3B). Interestingly, rotenone did not mitigate this effect of the hydroperoxide, a result consistent with the notion that the complex I inhibitor does not modify the extent of the insult received by the cells upon exposure to tB-OOH.

Rotenone or experimentally induced respiratory deficiency prevent the decline in cellular NPSH and ATP contents induced by tB-OOH. The capacity of suppression of electron transport to prevent cell death induced by tB-OOH may be explained by postulating that NADH will accumulate when oxygen is not being consumed by the respiratory chain. This will result in an enhanced availability of reducing equivalents, thus increasing the ability of the cells to recover from oxidative damage. The data illustrated in figure 4A are consistent with this premise because they indicate that rotenone markedly enhanced the NADH content of U937 cells. Furthermore, the NADH levels of respiration-deficient cells are markedly higher than those of the parent cell line (fig. 4B).


View larger version (26K):
[in this window]
[in a new window]
  		Fig. 4.   Rotenone, beta -OHB, pyruvate as well as experimentally induced respiratory deficiency enhances NADH levels. A, The cells were incubated in glucose-containing Saline A for 30 min in the absence or presence of either 0.5 µM rotenone or 5 mM pyruvate or 10 mM beta -OHB and then the NADH content was determined as detailed in "Materials and Methods." Results are expressed as the percent NADH content of sham-treated RP cells (1.63 ± 0.136 nmol/mg of protein) and represent the mean ± S.E.M. of three to six separate determinations. *P < .01 compared with RP cells by analysis of variance followed by Dunnett's test. B, NADH content of EB cells. Results are expressed as the percent NADH content of untreated RP cells and represent the mean ± S.E.M. of three separate determinations. *P < .01 compared with RP cells by analysis of variance followed by Dunnett's test.

It is known that treatment with tB-OOH reduces the NPSH pool (Geiger et al., 1993; Korytowski et al., 1995) and in fact the experiments illustrated in figure 5A indicate that a 30-min exposure to 1 mM tB-OOH induces a statistically significant decrease in cellular NPSH content. This effect was more pronounced upon treatment in the absence of glucose and was prevented by 0.5 µM rotenone, regardless of whether glucose was absent or present during treatment. Consistent with these results, exposure of EB or CP cells to 1 mM tB-OOH did not alter the NPSH pool (not shown).


View larger version (37K):
[in this window]
[in a new window]
  		Fig. 5.   Rotenone abolishes the decline in cellular NPSH and ATP promoted by tB-OOH. A, U937 cells were treated for 30 min with 1 mM tB-OOH in the absence or presence of rotenone in glucose-containing (open bars) or glucose-free (solid bars) Saline A. The NPSH levels were then determined as detailed in "Materials and Methods." Results represent the mean ± S.E.M. of three to seven separate determinations. *P < .05 and **P < .01 compared with appropriate controls by analysis of variance followed by Dunnett's test. B, U937 cells were exposed for 30 min to 0 (open bars) or 1 mM (striped bars) tB-OOH in glucose-containing Saline A in the absence or presence of 0.5 µM rotenone and then assayed for ATP content as detailed in "Materials and Methods." Results represent the mean ± S.E.M. of three separate determinations, each performed in duplicate. *P < .05 vs. untreated cells (unpaired t test).

Another consequence of exposure to tB-OOH is a decrease in the cellular ATP content. A number of laboratories have based the mechanism by which tB-OOH kills the cells on the drop in this adenine nucleotide (Nieminen et al., 1990, 1995; Imberti et al., 1993; Toussaint et al., 1994). The results illustrated in figure 5B indicate that a 30-min treatment in glucose-containing Saline A with 1 mM tB-OOH produced a small, although statistically significant, decline in ATP, and that this effect was prevented by 0.5 µM rotenone. A 3-hr postincubation in fresh culture medium allowed recovery of ATP to control values in cells that received treatment with tB-OOH (not shown). Rotenone alone did not significantly reduce ATP. In the absence of glucose, rotenone itself diminished the ATP pool and its effect was not additive to that exerted by tB-OOH (not shown).

NADH-linked substrates enhance oxygen consumption and prevent the toxicity of tB-OOH as well as the drop in cellular NPSH content. The results thus far presented are consistent with the possibility that inhibition of electron transport allows the maintenance of high levels of NADH which can then be utilized by the cell to recover from the insult inflicted by tB-OOH. To prove that this is indeed the case, we investigated the effect of two agents, namely pyruvate and beta -OHB, which enhance the intramitochondrial levels of NADH. The results illustrated in figure 4A demonstrate that indeed these two substrates, although less efficiently than rotenone, significantly augmented U937 cell NADH content. Under the same experimental conditions the NADH-linked substrates were found to increase the rate of oxygen consumption and this effect was inhibited by rotenone (fig. 6A). The cytotoxic effect of 1 mM tB-OOH in the absence or presence of 5 mM pyruvate or 10 mM beta -OHB is shown in figure 6C and the results indicate that both NADH-linked substrates, while not being toxic for the cells (not shown), virtually abolished the cytotoxicity of tB-OOH. Finally, we investigated the effect of the NADH-linked substrates on the tB-OOH-induced decrease in cellular NPSH. The results illustrated in figure 6D indicate that this effect was prevented under conditions in which the exposure to tB-OOH was performed in the presence of either 5 mM pyruvate or 10 mM beta -OHB.


View larger version (36K):
[in this window]
[in a new window]
  		Fig. 6.   Pyruvate as well as beta -OHB enhances U937 cell oxygen consumption and prevents the cytotoxic response and decline in NPSH levels promoted by tB-OOH. A, U937 cell oxygen consumption was measured in glucose-containing Saline A after addition of beta -OHB (10 mM) or pyruvate (5 mM) and further addition of rotenone (0.5 µM). B,U937 cells were exposed to 1 mM tB-OOH in glucose-containing Saline A in the absence (open circles) or presence (closed circles) of 5 mM pyruvate. The levels of tB-OOH were measured in the extracellular milieu at the indicated time points, as detailed in "Materials and Methods." Results represent the mean ± S.E.M. of three separate determinations, each performed in duplicate. The inset shows the extent of GPx-inhibition induced by tB-OOH (1 mM for 15 min) in the absence or presence of pyruvate (mean values from two separate experiments with similar outcomes). C, U937 cells were preexposed for 5 min in glucose containing Saline A to either of beta -OHB (10 mM) or pyruvate (5 mM) and then treated for further 30 min with 1 mM tB-OOH in the absence (a) or presence (b) of the NADH-linked substrates. Cells were analyzed for cytotoxicity after a 6-hr posttreatment incubation in fresh culture medium. Neither beta -OHB nor pyruvate were cytotoxic when given alone to the cultures. Experimental results represent the mean ± S.E.M. of at least three independent determinations, each performed in duplicate. *P < .01 compared with cells treated with tB-OOH alone by analysis of variance followed by Dunnett's test. D, 937 cells were treated for 30 min as described in (C, condition b) and the NPSH content was immediately determined, as detailed in "Materials and Methods." Experimental results represent the mean ± S.E.M. of three to six independent determinations. *P < .01 compared with untreated cells by analysis of variance followed by Dunnett's test.

It is important to note that previous studies (Marcengill et al., 1995) had reported that tB-OOH promotes non-enzymatic decarboxylation of pyruvate, an event which may explain the protective effects afforded by this NADH-linked substrate. Under the conditions used in our experiments, however, we did not find evidence of chemical interaction between the hydroperoxide and the NADH-linked substrates. This inference is supported by the observation that pyruvate neither affected the rate of disappearance of tB-OOH from the culture medium (fig. 6B) nor the extent of tB-OOH-induced GPx inhibition (inset to fig. 6B). Finally, pyruvate as well as beta -OHB were cytoprotectants also using a treatment protocol involving a 5-min exposure to the NADH-linked substrates followed by challenge (30 min) with tB-OOH in their absence (fig. 6C). Further experimental evidence in support of the above inference is provided by the results shown in figure 7 (see below).


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 7.   Pyruvate or rotenone are not cytoprotective for GSH-depleted cells treated with tB-OOH. After incubation for 16 hr in complete culture medium containing 10 µM BSO, the cells were treated for 30 min with increasing concentrations of tB-OOH in the absence (circles) or presence of 5 mM pyruvate (triangles) or 0.5 µM rotenone (squares). Pyruvate and rotenone were added 5 min before tB-OOH. Treatments were performed in glucose-containing Saline A at 37°C and were followed by a 6-hr postincubation in drug-free culture medium. Toxicity was assessed using the trypan blue exclusion assay. Results represent the mean ± S.E.M. calculated from at least three separate experiments, each performed in duplicate.

Rotenone or pyruvate does not prevent the toxicity of tB-OOH in NPSH-depleted cells. Cellular NPSH levels can be easily manipulated by culturing the cells in the presence of BSO, an inhibitor of the synthesis of GSH, the major constituent of the NPSH pool. Indeed, an overnight exposure of U937 cells to 10 µM BSO resulted in an 85% reduction of the NPSH content. As illustrated in figure 7, thiol-depleted cells were extremely sensitive to the killing elicited by tB-OOH and this response was not affected by 0.5 µM rotenone or by 5 mM pyruvate.

Rotenone or pyruvate prevents the loss of mitochondrial membrane potential induced by tB-OOH. A number of studies have provided a detailed description of the effects of tB-OOH on mitochondria (Masaki et al., 1989b; Wu et al., 1990; Livingston et al., 1992; Imberti et al., 1993; Castilho et al., 1995) and it is now well accepted that collapse of the mitochondrial membrane potential and opening of permeability transition pores in the inner mitochondrial membrane is the final event leading to cell death (Castilho et al., 1995). Indirect evidence that this mechanism is involved also under the experimental conditions utilized in this study is given by the results illustrated in figure 8B showing that the onset of cell death is associated with, and preceded by, loss of rhodamine 123 staining (compare with A, untreated cells) and that both events were fully prevented by 0.5 µM cyclosporin A (E). Interestingly, pyruvate (C) and rotenone (D) displayed similar protective effects.


View larger version (40K):
[in this window]
[in a new window]
  		Fig. 8.   Cyclosporin A as well as pyruvate or rotenone prevent tB-OOH-induced mitochondrial depolarization and cytotoxicity. U937 cells (2 × 105/ml) were treated for 5 min with 5 mM pyruvate or 0.5 µM rotenone or for 15 min with 0.5 µM cyclosporin A and then treated with 1 mM tB-OOH for a further 30 min. The cells were then washed and resuspended in fresh culture medium, and incubated at 37°C for 4.5 hr 30 min before the end of the posttreatment incubation, rhodamine 123 (10 µg ml) was added to the cultures. Rhodamine 123 is taken up by the mitochondria and retained as a function of the electrochemical and proton gradient. As the potential dissipates the marker is progressively excluded. The cells were analyzed for cell killing (by trypan blue exclusion assay, see Materials and Methods") and for mitochondrial depolarization using fluorescence microscopy. Experimental results represent the mean ± S.E.M. of at least three independent determinations. *P < .01 compared with cells treated with tB-OOH alone by analysis of variance followed by Dunnett's test. A, control; B, 1 mM tB-OOH; C, 1 mM tB-OOH + 5 mM pyruvate; D, 1 mM tB-OOH + .5 µM rotenone; E, 1 mM tB-OOH + 0.5 µM cyclosporin A.

    Discussion
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
Conclusions
References

Five principal findings emerge from this work. First, two distinct experimental approaches allowing the blockage of electron transport in the respiratory chain, namely the complex I inhibitor rotenone or experimentally induced respiratory deficiency, prevented U937 cell death promoted by tB-OOH. Second, NADH-linked substrates were also efficient cytoprotectants in tB-OOH-injured cells. Third, suppression of electron transport as well as supplementation with NADH-linked substrates enhanced mitochondrial NADH and prevented the decline in GSH and ATP elicited by tB-OOH. Fourth, accumulation of mitochondrial NADH did not afford protection in GSH-depleted cells treated with tB-OOH. Finally, accumulation of mitochondrial NADH prevented the loss of mitochondrial membrane potential and the opening of permeability transition pores caused by tB-OOH.

Taken together these findings strongly suggest that depletion of mitochondrial NADH mediated by the respiratory chain is a critical event in the toxic response of cultured U937 cells to challenge with tB-OOH.

Chemically or experimentally induced respiratory deficiency promotes survival of tB-OOH-injured U937 cells. The following observations provide direct experimental evidence that electron transport in the respiratory chain is detrimental for tB-OOH-injured cells.

Rotenone prevented cell death induced by tB-OOH (figs. 1B and C) and this effect appeared to be specifically linked to its ability to block electron transport at the level of complex I. Indeed, rotenone afforded cytoprotection over the same concentration-range in which it inhibited oxygen consumption (inset to fig. 1C). Plotting the concentration-response curves of rotenone-induced inhibition of oxygen utilization and of cell death caused by tB-OOH revealed an important correlation between these two parameters.

The possible involvement of other mechanisms such as iron chelation, radical- scavenging effects or alterations in peroxide metabolism were ruled out by the observations that rotenone 1) did not inhibit the toxicity of hydrogen peroxide (inset to fig. 1B), 2) did not affect DNA single strand breakage induced by tB-OOH (fig. 3A), 3) did not affect the rate of disappearance of tB-OOH from the culture medium (fig. 3B), 4) did not modify the extent of GPx inhibition caused by tB-OOH (inset to fig. 3B) and 5) did not reduce the toxicity of tB-OOH in respiration-deficient cells (fig. 2) or respiration-proficient, GSH-depleted cells (fig. 7).

Cells that were made respiration-deficient on growth in either ethidium bromide (which interferes with mitochondrial DNA and RNA synthesis) or chloramphenicol (which abolishes mitochondrial protein synthesis) were highly resistant to cell death promoted by tB-OOH (fig. 2). The specificity of this resistance is emphasized by the observation that these cells were as sensitive as respiration-proficient cells to the toxicity of hydrogen peroxide (inset to fig. 2) or to DNA cleavage induced by tB-OOH (fig. 3A). Furthermore, the rate of tB-OOH metabolism in the three cell types was basically identical (fig. 3B). The superimposable toxicity curves observed in respiration-deficient cells and rotenone-supplemented respiration-proficient cells in response to 1 to 6 mM levels of tB-OOH are noteworthy.

These data strongly support the idea that electron transport in the respiratory chain is highly detrimental for cells injured with tB-OOH.

Electron transport promotes killing of tB-OOH-injured U937 cells via wasteful consumption of NADH. Pyruvate or beta -OHB, which augment intramitochondrial NADH (Khan and O'Brien, 1995) (fig. 4A) and stimulate oxygen consumption (fig. 6A), also prevented the lethal effects of tB-OOH in cultured U937 cells (fig. 6C). These effects were not dependent on tB-OOH-induced nonenzymatic decarboxylation of the NADH-linked substrates (Marcengill et al., 1995) for a number of reasons. First, the rate of disappearance of tB-OOH from the culture medium was not affected by pyruvate (fig. 6B). Second, pyruvate did not modify the extent of GPx inhibition induced by tB-OOH (inset to fig. 6B). Third, pyruvate did not reduce the toxicity of micromolar levels of tB-OOH in GSH-depleted cells (fig. 7). Finally, previous results from our laboratory (Guidarelli et al., 1996b) demonstrated that DNA single strand breakage induced by tB-OOH was not decreased but actually augmented by pyruvate.

Taken together these results provide solid experimental evidence supporting the above premise and allow us to conclude that the cytoprotective effects afforded by the NADH-linked substrates are not the consequence of depletion of the peroxide mediated by their chemical interaction.

Thus, prevention of tB-OOH-induced cell death can be achieved both via suppression of electron transport, a condition resulting in accumulation of mitochondrial NADH, and via stimulation of electron transport mediated by NADH-linked substrates. As a consequence, the mechanism whereby electron transport affects the toxic response to tB-OOH appears to be uniquely dependent on depletion of the mitochondrial fraction of this pyridine nucleotide. Consistent with this notion are the observations that 1) rotenone augments mitochondrial NADH (fig. 4A) and 2) respiration-deficient cells display markedly higher NADH content than respiration-proficient cells (fig. 4B).

Thus, mitochondrial NADH appears to serve a pivotal role in determining the cellular outcome after tB-OOH exposure. This is in contrast with the results obtained with H2O2 since the toxicity brought about by this treatment was insensitive to rotenone and glucose (fig. 2B and inset) as well as to the respiration-deficient phenotype (inset to fig. 2). As a consequence, mitochondrial NADH does not appear to be critical in the cytotoxic response evoked by H2O2.

NADH can be used in the reductive biosynthesis (via interconversion to NADPH) and for the energy production that are needed for efficient recovery from the damage inflicted by the hydroperoxide. In this study we provide experimental evidence indicating that accumulation of mitochondrial NADH, achieved via suppression of electron transport or supplementation of NADH-linked substrates, prevents the decline in cellular NPSH induced by tB-OOH (figs. 5A and 6D). Under the same conditions NADPH was also found to increase (L. Palomba, unpublished data). The fact that these effects were associated with prevention of cell death strongly suggests the existence of a cause-effect relationship, a conclusion further supported by the observation that the tB-OOH-induced cytotoxic response was not inhibited in NPSH-depleted cells (fig. 7). These results imply that mitochondrial NADH is primarily involved in maintaining thiols in a reduced state. Although we did not attempt to distinguish between cytosolic and mitochondrial NPSH, it is to be expected that mitochondrial pyridine nucleotides will primarily prevent depletion of mitochondrial NPSH. Thus, our results indirectly suggest that, under the experimental conditions utilized in this study, tB-OOH mainly affects the mitochondrial NPSH pool. The fact that prevention of this effect is beneficial in the cytotoxic response to tB-OOH, is emphasized by the results obtained in a recent study (Shan et al., 1993) indicating that selective depletion of mitochondrial, but not cytosolic, GSH markedly enhances the toxicity of the hydroperoxide.

The oxidation of GSH leads to formation of GSSG and the reduction of the latter compound is catalyzed by the enzyme glutathione reductase. The fact that this enzyme is essentially specific for NADPH (Kehrer and Lund, 1994) would therefore suggest that NADH provides the reducing equivalents for formation of NADPH. Indeed, a previous study (Kurosawa et al., 1990) demonstrated that NADPH generated by transhydrogenation accounted for at least 50% of the reduction of GSSG generated by tB-OOH in isolated mitochondria. It is important to note, however, that the biochemical events leading to the formation of pyridine nucleotides are closely linked to energy production (Kehrer and Lund, 1994). Thus, the question arises as to whether prevention of tB-OOH-induced cell death mediated by mitochondrial NADH is uniquely dependent on the supply of reducing equivalents or is also the consequence of effects on energy production. The second possibility finds apparent experimental support in the results depicted in figure 5B showing that the small decline in cellular ATP caused by tB-OOH was prevented by rotenone (fig. 5B). Under the experimental conditions used in study we observed that in the presence of glucose (fig. 5B), unlike in its absence (not shown), rotenone did not reduce ATP levels, thus suggesting that the glycolytic pathway, in the absence of mitochondrial ATP synthesis, provides sufficient amounts of the nucleotide during the 30-min exposure. The fact that rotenone prevented the effects of tB-OOH on ATP is therefore consistent with the possibilities that 1) tB-OOH inhibits nonmitochondrial ATP synthesis and that 2) this block can be indirectly removed by mitochondrial NADH. Several mechanisms by which reducing equivalents are transferred from the mitochondria to the cytosol have been described (Kehrer and Lund, 1994).

Taken together, these results would therefore suggest that mitochondrial NADH provides the reducing equivalents necessary for recovery from the insult inflicted by tB-OOH. Prevention of NPSH depletion appears to be one of these effects. Because tB-OOH caused a small reduction in the ATP levels, it is difficult to establish whether this effect has any impact at all in terms of cell viability. As a consequence, it appears unlikely that cytoprotection afforded by rotenone (or pyruvate) is dependent on effects on ATP.

Our conclusion is therefore in apparent contrast with that of other authors who have suggested that the toxicity of tB-OOH is dependent on depletion of cellular ATP (Imberti et al., 1993). This conclusion was mainly based on the observation that supply of the glycolytic substrate fructose reduces the killing of hepatocytes promoted by tB-OOH. Stimulation of the glycolytic pathway, however, will also enhance mitochondrial NADH. As a final note, it is important to emphasize that the results discussed above indicate that in U937 cells ATP is effectively generated by the glycolytic pathway. The possibility therefore exists that ATP depletion might be of greater importance in those cells in which oxidative phosphorylation is critical in maintenance of cellular ATP levels.

Mitochondrial NADH prevents the loss of mitochondrial membrane potential. Previous work from Nieminen et al. (1995) demonstrated that the onset of cell death brought about by tB-OOH is preceded by and causally linked to mitochondrial permeability transition and depolarization. Our results indicating that cyclosporin A suppressed the leakage of rhodamine 123 from the mitochondria of cells treated with tB-OOH as well as the lethal effects mediated by the hydroperoxide (fig. 8E), would suggest that also under the experimental conditions utilized in the present study mitochondrial membrane permeability transition and depolarization are important events in the cytotoxic response to the hydroperoxide. Indeed, the above immunosuppressive drug was previously shown to bind to cyclophillin thereby preventing pore opening (Halestrap and Davidson, 1990). The fact that both rotenone (fig. 8D) and pyruvate (fig. 8C) mimicked the effects of cyclosporin A is therefore consistent with the notion that an enhancement in intramitochondrial NADH prevents the deleterious effects of the hydroperoxide at the mitochondrial level. Although this study did not directly test or demonstrate a causal role for mitochondrial damage in cell death evoked by tB-OOH, the above results reveal an important correlation between these two parameters.

    Conclusions
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
Conclusions
References

In this study we provide experimental evidence indicating that mitochondrial NADH serves a pivotal role for determining the outcome of a short term exposure of U937 cells to a high concentration of tB-OOH. This pyridine nucleotide would appear to be able to prevent depletion of NPSH and counteract the sequence of events leading to mitochondrial dysfunction and cell death. Electron transport in the mitochondrial respiratory chain at the time of tB-OOH exposure is therefore detrimental because it leads to wasteful consumption of mitochondrial NADH. Our results, while confirming the emerging theory that mitochondrial damage is the cause of tB-OOH-induced cell death, identify a critical upstream event in this response. Supplementation of NADH-linked substrates may represent a therapeutic strategy alternative to antioxidants for counteracting the deleterious effects of an acute oxidative stress.

    Footnotes

Accepted for publication November 21, 1997.

Received for publication July 2, 1997.

Send reprint requests to: Prof. Orazio Cantoni, Istituto di Farmacologia e Farmacognosia, Università di Urbino, Via S. Chiara, 27, 61029 Urbino (PS), Italy.

    Abbreviations

tB-OOH, tert-butylhydroperoxide; BSO, L-buthionine-[S, R]-sulphoximine; beta -OHB, beta -hydroxybutyrate; NPSH, non-protein sulphydryl; GPx, glutathione peroxidase; GSH, reduced glutathione; NADH, nicotinamide adenine dinucleotide reduced form.

    References
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
Conclusions
References

0022-3565/98/2843-1112$03.00/0
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright © 1998 by The American Society for Pharmacology and Experimental Therapeutics


This article has been cited by other articles:


Home page
 Am. J. Physiol. Cell Physiol.Home page
D. D. Arrington and R. G. Schnellmann
Targeting of the molecular chaperone oxygen-regulated protein 150 (ORP150) to mitochondria and its induction by cellular stress
Am J Physiol Cell Physiol, February 1, 2008; 294(2): C641 - C650.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow