Oxidative stress has been associated with multiple pathologies and disease states, including those involving the cardiovascular system. Previously, we showed that pulmonary artery endothelial cells (PAECs) undergo apoptosis after acute exposure to H2O2. However, the underlying mechanisms regulating this process remain unclear. Because of the prevalence of H2O2 in normal physiological processes and apparent loss of regulation in disease states, the purpose of this study was to develop a more complete understanding of H2O2-mediated adverse effects on endothelial cell survival. Acute exposure of PAECs to H2O2 caused a dose-dependent increase in cellular release of lactate dehydrogenase and a significant increase in production of superoxide ions, which appear to be generated within the mitochondria, as well as a significant loss of mitochondrial membrane potential and activity. Subsequent to the loss of mitochondrial membrane potential, PAECs exhibited significant caspase activation and apoptotic nuclei. We also observed a significant increase in intracellular free Zn2+ after bolus exposure to H2O2. To determine whether this increase in Zn2+ was involved in the apoptotic pathway induced by acute H2O2 exposure, we developed an adenoviral construct for overexpression of the Zn2+-binding protein metallothionein-1. Our data indicate that chelating Zn2+, either pharmacologically with N,N,N′,N-tetrakis(2-pyridylmethyl)ethylene diamine or by overexpression of the Zn2+-binding protein metallothionein-1, in PAECs conferred significant protection from induction of apoptosis and cell death associated with the effects of acute H2O2 exposure. Our results show that the acute toxicity profile of H2O2 can be attributed, at least in part, to liberation of Zn2+ within PAECs. We speculate that regulation of Zn2+ levels may represent a potential therapeutic target for cardiovascular disease associated with acute oxidative stress.
- oxidant stress
- cell signaling/signal transduction
an increasing body of research indicates that oxidative stress underlies a number of disease states, including hypertension, diabetes, and ischemic insult resulting from stroke, atherosclerosis, and aging (3, 17, 35, 54, 59). In normal physiological conditions, exposure to reactive oxygen species (ROS) is an inevitable consequence of aerobic metabolism. ROS result from the partial reduction of molecular oxygen [i.e., superoxide anions (O2•−), hydroxyl radicals (·OH), and H2O2] and are more highly reactive than molecular oxygen. ROS can be generated from several physiological situations and account for ∼2–4% of total molecular oxygen consumed in normal aerobic metabolism (48). ROS can also act as second messengers in several signal transduction pathways (26, 46, 52) and as nonspecific defenses in immune responses (14, 18).
It is becoming apparent that H2O2 has an important and variable role in mammalian cell physiology. Under normal physiological conditions, most intracellular H2O2 is formed by the dismutation of O2•−, a by-product of incomplete reduction of oxygen during oxidative phosphorylation. Evidence indicates that H2O2 can function as a second messenger in multiple signal transduction pathways, in a manner similar to nitric oxide (NO). A number of different cell types have been shown to produce O2•− and H2O2 in response to extracellular stimuli (47). H2O2 concentrations in tissues and fluids in the human body vary widely (20), and the sensitivity of different cells varies according to intracellular iron content and/or the presence of enzymes such as catalases and glutathione peroxidases and systems associated with thioredoxin (9). Concentrations in the hundreds of micromolar have been reported, and consumption of common beverages and smoking may result in exposure levels of >500 μM (21). Loss of H2O2 regulation, from excessive production or compromised ability to dispose of H2O2, is believed to play a central role in disease and aging (48), and elevated levels of H2O2 in expired air have been reported in patients suffering from airway inflammation (30). Indeed, we reported that acute exposure to high levels of H2O2 induced an apoptotic response in pulmonary endothelial cells (58). When H2O2 reaches potentially toxic levels, most attention has focused on the oxidative damage to proteins, lipid, and nucleic acids. However, because perhaps 30–50% of proteins in a given proteome of a cell contain Zn2+-binding domains, there is the distinct possibility that H2O2-mediated toxicity alters cellular Zn2+ homeostasis, specifically, the release of Zn2+ due to oxidative modification of Zn2+-containing proteins. Therefore, in this study, we investigated the potential mechanistic role of alterations in Zn2+ homeostasis in endothelial cells acutely exposed to H2O2. We report that increased intracellular Zn2+ is a key event associated with the disruption of mitochondrial function and the induction of the apoptotic pathway by acute H2O2-mediated oxidative stress in pulmonary artery endothelial cells (PAECs). Furthermore, we report that adenoviral-mediated overexpression of the Zn2+-sequestering protein metallothionein (MT)-1 in endothelial cells resulted in resistance to H2O2-induced apoptosis. Although other cytoprotective functions of MT-1 may be involved, our data suggest that the sequestration of intracellular free Zn2+ may represent a potential therapeutic target for mitigation of the acute oxidative stress underlying many cardiovascular disease states.
MATERIALS AND METHODS
Primary cultures of ovine fetal PAECs were isolated as described previously (57). Endothelial cell identity was confirmed by their typical cobblestone appearance, contact inhibition, specific uptake of 1,1′-dioctadecyl-1,3,3′,3′-tetramethylindocarbocyanine perchlorate-labeled acetylated low-density lipoprotein (Molecular Probes, Eugene, OR), and positive staining for von Willebrand factor (Dako, Carpentaria, CA). Cells were maintained in DMEM containing phenol red (MediaTech, Herndon, VA) supplemented with 10% FCS (Hyclone, Logan, UT) and antibiotic-antimycotic (MediaTech) at 37°C in a humidified atmosphere of 5% CO2-95% air. Cells were between passages 3 and 10, seeded at ∼50% confluence, and utilized when fully confluent. Before experimental treatment, except as noted, cells were trypsinized, counted with a hemocytometer, and replated in 6-, 24-, or 96-well plates (Costar, Corning, NY) at a density of 5 × 105 cells/cm2 and allowed to adhere for ≥18 h. Cells were then induced to quiescence by serum starvation with replacement of normal DMEM with serum-free, phenol red-free DMEM (GIBCO/BRL, Gaithersburg, MD) and allowed to incubate overnight.
Exposure to H2O2.
Intracellular levels of H2O2 were determined by fluorescence microscopy (see below), which measured oxidation of the nonfluorescent 2′,7′-dichlorofluorescein (DCFH, Sigma) to the fluorescent 2′,7′-dichlorofluorescein (DCF) at 485-nm excitation and 530-nm emission. H2O2 concentrations were selected to span the range of concentrations that have been observed in human tissues (20) and also to encompass the observed phenomenon in this particular cell line, which shows greater resistance than cell lines used in other reports (1, 9). PAECs exposed to 0–1 mM H2O2 for 4 h in serum-free, phenol red-free DMEM were measured at 0, 30, 60, 120, and 240 min. At 240 min, all experimental groups had returned to within 10% of endogenous fluorescence levels. Mean intracellular fluorescence was plotted for each experimental group, and overall H2O2 exposure was estimated using the trapezoidal rule for standard area under the curve, as previously described (44). Exposure levels were calculated to be ∼3-, 15-, 40-, and 55-fold vs. untreated for 100, 250, 500, and 1,000 μM, respectively.
A personal computer-based imaging system consisting of an Olympus IX51 microscope equipped with a charge-coupled device camera (Hamamatsu Photonics, Hamamatsu City, Japan) was used for acquisition of fluorescent images. Fluorescent-stained cells were observed with the appropriate excitation and emission, and the average fluorescent intensities (to correct for differences in cell number) were quantified using ImagePro Plus imaging software (version 5.0, Media Cybernetics, Silver Spring, MD). Statistical analysis of difference between treatments was carried out as described in Statistical analysis.
Analysis of cell death.
Quiescent PAECs were cultured in 24-well plates and treated with H2O2 (see above). After incubation, the medium was collected and centrifuged for 5 min at 500 g, and the supernatant was stored at 4°C until assay. Relative cytotoxicity was quantified by measurement of release of the soluble cytoplasmic enzyme lactate dehydrogenase (LDH). LDH activity in cell-free supernatant was measured using a commercial kit (Roche Applied Science, Indianapolis, IN). After a given time course, 50-μl aliquots of cell-free medium were transferred from all wells to a fresh 96-well plate, and 50 μl of reconstituted substrate mix were added to each well. The plate was incubated for 30 min at room temperature and protected from light. The reaction was halted by addition of 50 μl of stop solution to each well, and the absorbance was recorded at 490 nm. Each culture sample was measured in quadruplicate, with a minimum of three samples per experimental group. Relative cytotoxicity was determined by comparison of absorbance of the experimental group with absorbance of a control cell group treated with 1% Triton X-100 cell lysis buffer according to the manufacturer’s protocol.
Determination of apoptotic events.
PAECs were seeded onto 96-well plates and treated as described above. Caspase activation was visualized by cotreatment of cells with 1 μM CaspACE FITC-VAD-FMK In Situ Marker (Promega, Madison, WI), a fluorescent analog of the pan-caspase inhibitor N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone, which readily enters cells and binds irreversibly to activated caspases (5). Fluorescent cells were observed at 485-nm excitation and 530-nm emission. After treatment, cells were washed with H2O2-free DMEM and then incubated overnight in phenol red-free medium with 10% FCS. At 18 h after the onset of exposure, cells were incubated in medium with 10 μM FITC-conjugated caspase inhibitor (FITC-Val-Ala-Asp-fluoromethyl ketone). After 20 min of incubation at 37°C in the dark, cells were washed with fresh medium and visualized using fluorescence microscopy.
A second method for determination of apoptotic induction involved TdT-mediated dUTP nick end labeling (TUNEL). PAECs treated as described above were incubated for ∼18 h in phenol red-free DMEM supplemented with 10% FBS. After incubation, analysis was performed as we previously described (56). TUNEL-positive cells were visualized by indirect immunofluorescence with 485-nm excitation and 530-nm emission, and total cell number was calculated from propidium iodide counterstaining and visualization at 535-nm excitation and 617-nm emission.
Analysis of mitochondrial function.
Quiescent PAECs were cultured in 96-well plates and treated with H2O2 (see above), and mitochondrial function was quantified using a tetrazolium assay (CellTiter 96 AQueous MTS Reagent, Promega) according to the manufacturer’s instructions. The tetrazolium reagent is bioreduced to a colored product, the quantity of which is proportional to the number of metabolically active cells. Reagent (20 μl) was added directly to cells in 100 μl of medium, and after 4 h of incubation at 37°C the absorbance at 492 nm was read using a plate reader (Labsystems Multiskan EX, Fisher, Hampton, NH). For the proliferation assay, cells were washed with PBS and incubated in medium containing 10% FCS and H2O2 at 0- to 55-fold vs. untreated. To account for variances in cell number that might affect the strength of the signal, control wells were fixed, permeabilized, and stained with propidium iodide, and total cell number was calculated from propidium iodide counterstaining and visualization at 535-nm excitation and 617-nm emission. Signal strength was then normalized to cell number.
Mitochondrial membrane potential (Ψm) has been analyzed previously using the lipophilic cation 5,5′6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine iodide (11), which fluoresces red in its multimeric form in healthy mitochondria and is the active reagent in the DePsipher mitochondrial potential assay kit (Trevigen, Gaithersburg, MD). PAECs were seeded onto 96-well plates and incubated with 0- to 55-fold H2O2 (see above). DePsipher reagent (25 μg/ml) was added after treatment with H2O2, and the sample was incubated for a further 20 min. After an additional wash with Dulbecco’s PBS (DPBS), the cytosolic monomer (green) form was observed and quantified by fluorescence microscopy at 530 nm (see above).
Mitochondrial function can also be analyzed by cardiolipin, which can be detected by staining with 10-N-nonyl acridine orange (NAO; Sigma-Aldrich, St. Louis, MO) dye. Mitochondrial disruption results in release of cardiolipin from the inner mitochondrial leaflet into the cytosol. On binding with the cytosolic monomer form of cardiolipin, NAO will fluoresce at 510-nm excitation and 580-nm emission. Quiescent PAECs were plated into 24-well plates and treated as previously described. NAO (10 mM in H2O, diluted 1:1,000) was added to the medium 30 min before the end of the experiment. Cells were washed thoroughly with PBS and imaged using fluorescence microscopy (see above).
Measurement of cellular O2•− levels.
Intracellular O2•− production was measured using two methods: the cell-permeable dye dihydroethidium (DHE; Molecular Probes), which binds to nuclear DNA when oxidized by O2•− and emits red fluorescence (19), and electron paramagnetic resonance (EPR) spectroscopy with the spin probe 1-hydroxy-3-methoxycarbonyl-2,2,5,5-tetramethylpyrrolidine·HCl (CMH; Alexis Biochemicals, San Diego, CA). For the DHE analysis, after H2O2 exposure, cells were immediately incubated with 10 μM DHE for 10 min at 37°C. Cells were then washed in fresh DHE-free medium, and the fluorescence intensity was quantified using a personal computer-based imaging system consisting of an Olympus IX51 microscope equipped with a charge-coupled device camera (Hamamatsu Photonics). DHE-stained cells were observed after 518-nm excitation and 605-nm emission, and the average fluorescent intensities (to correct for differences in cell number) were quantified using ImagePro imaging software (Media Cybernetics). Statistical analyses between treatments were carried out as described in Statistical analysis.
The EPR analysis was carried out using cells treated with or without polyethylene glycol-conjugated superoxide dismutase (PEG-SOD, 100 U/ml; Sigma) to quantify the SOD-inhibitable formation of CM·. Cells were plated in six-well plates and treated as described above. In the final hour of incubation with H2O2, 20 μl of spin-trap stock solution consisting of CMH [20 μM in DPBS + 25 μM desferrioxamine (Sigma-Aldrich) and 5 μM diethyldithiocarbamate] and 2 μl of DMSO were added to each well. On completion of incubation, the medium was removed and the adherent cells were trypsinized and pelleted at 500 g. The cell pellet was washed and suspended in a final volume of 35 μl of DPBS + desferrioxamine and diethyldithiocarbamate, loaded into a 50-μl capillary tube, and analyzed with a MiniScope MS200 ESR (Magnettech, Berlin, Germany) at 40-mW microwave power, 3,000-mG modulation amplitude, and 100-kHz modulation frequency. EPR spectra were analyzed and measured for amplitude using ANALYSIS version 2.02 software (Magnettech), and experimental groups were compared as described in Statistical analysis.
Detection of mitochondrial O2•− production.
Quiescent PAECs were plated onto 24-well plates and treated as previously described. Mitochondrial O2•− production was measured using MitoSOX Red mitochondrial O2•− indicator (Molecular Probes), a fluorogenic dye for selective detection of O2•− in the mitochondria of live cells. MitoSOX Red reagent is live-cell permeant and is rapidly and selectively targeted to the mitochondria. Once in the mitochondria, MitoSOX Red reagent is oxidized by O2•− and exhibits bright red fluorescence on binding to nucleic acids. Briefly, after treatment with H2O2, cells were washed with fresh medium and then incubated in medium containing 2 μM MitoSOX Red for ∼10 min at 37°C in the dark. Cells were washed with fresh serum-free medium and imaged using fluorescence microscopy at 510-nm excitation and 580-nm emission.
Generation of MT-1 adenoviral expression construct.
An adenoviral vector expressing MT-1 was constructed with the pAd/pENTR system (Invitrogen). Briefly, a pAd/CMV plasmid containing an MT-1 cDNA (cDNA purchased from American Type Culture Collection, Manassas, VA) was created and transfected into 293A cells. A selected clone was propagated, and viral lysate was collected and titered according to the manufacturer’s protocol. The titer for the AdV.MT-1 preparation was 3.4 × 1010 plaque-forming units/ml. PAECs were infected for 120 min with virus diluted in normal DMEM to the desired multiplicity of infection. Expression was verified by Western blot analysis.
Pharmacological Zn2+ chelation.
As an alternative to MT-1 protein-mediated sequestration/chelation of Zn2+, a chemical chelation agent, with N,N,N′,N-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN; Calbiochem), was utilized. We previously determined that 6.5 μM is the maximum concentration that does not induce significant toxicity in this cell line (60).
Western blot analysis.
After treatment, cells were harvested via trypsinization, washed twice with PBS, assayed for protein content by Bio-Rad protein assay (Bio-Rad, Hercules, CA), resuspended in 2× Laemmli protein sample buffer, and boiled for ∼5 min. Approximately 20–40 μg of each protein sample were loaded and subjected to SDS-PAGE on a 4–20% Tris-SDS-HEPES gel (Gradipore, Frenchs Forest, Australia) and electrophoretic transfer to a polyvinylidene difluoride membrane (Bio-Rad), which was blocked overnight in 5% nonfat dry milk in TBS + Tween 20 with gentle shaking at 4°C. Membranes were probed with anti-MT-1 antibody (QED, Malden, MA) and then with a horseradish peroxidase-conjugated secondary antibody. After visualization of MT-1 protein bands, membranes were stripped for 30 min at room temperature with stripping buffer [25 mM glycine-HCl and 1% SDS (pH 2.0)], washed in PBS, blocked, and probed as described above using anti-β-actin antibody (Sigma). Protein bands were visualized using chemiluminescence (SuperSignal West Femto Substrate Kit, Pierce, Rockford, IL) on a Kodak 440CF image station (Kodak, Rochester, NY). Band intensity was quantified using Kodak 1D image processing software, and signal for MT-1 was normalized for differential protein loading using signal detected for β-actin.
Quantification of intracellular Zn2+ levels.
Cells were plated in 24-well plates, treated as described above, and loaded with 2.5 μM FluoZin 3-AM (Molecular Probes), a fluorescent dye specific for unassociated Zn2+ (16), at 37°C for 30 min before the end of the experimental time course. Cells were washed twice with serum-free, phenol red-free DMEM and incubated for an additional 30 min in the dark to complete the intracellular cleavage of the AM ester. Samples were analyzed by fluorescence microscopy as described above at 494-nm excitation and 516-nm emission. A similar procedure was used with another Zn2+-specific fluorophore, zinquin ethyl ester (ethyl[[2-methyl-8-[[(4-methylphenyl)sulfonyl]amino]-6-quinolinyl]oxy] acetate; Biotium, Hayward, CA) (62). Briefly, after treatment, cells were washed with DPBS and incubated with 20 μM zinquin in DPBS for 30 min at room temperature, washed with DPBS, and imaged using fluorescence microscopy at 368-nm excitation and 490-nm emission.
Values are means ± SE from at least three experiments. Statistical analysis between experimental groups was performed by one-way ANOVA. The statistical significance of differences was set at P < 0.05.
There is some discrepancy in the literature concerning the concentrations of H2O2 that represent an accurate physiological representation, especially in situations of acute oxidative stress. Therefore, we utilized concentrations that span those previously reported in human tissues and in vitro (20). Thus PAECs were exposed to 0–1 mM H2O2 for 4 h in serum-free, phenol red-free DMEM, and maximum overall H2O2 exposure was estimated by DCF-diacetate oxidation to estimate intracellular H2O2 levels. From our preliminary studies, maximal exposure levels were calculated to be ∼3-, 15-, 40-, and 55-fold vs. untreated for 100, 250, 500, and 1,000 μM H2O2, respectively.
Our initial results indicated a dose- and time-dependent increase in LDH release in cells exposed to 0–1 mM H2O2 for 4 h (Fig. 1, A and B) in the absence of any loss in cell number or density on the plates as observed in bright field or with propidium iodide staining (data not shown). Results similar to those of LDH release were found using 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium assay for mitochondrial activity, indicating that alterations in mitochondrial function underlie, at least in part, the H2O2-mediated toxic effects on PAECs (Fig. 1C). In addition, we confirmed our previously published findings, inasmuch as PAECs showed a significant increase in immunohistochemical staining for activated caspases (Fig. 1D) and TUNEL-positive cells (Fig. 1E) after exposure to ≥500 μM H2O2. Using DePsipher assay to determine the state of Ψm, we also confirmed that the apoptotic pathway was associated with a loss of mitochondrial function. After exposure to H2O2, we detected a significant loss of Ψm (Fig. 2A). In addition, staining with NAO showed a significant increase in monomeric cardiolipin (Fig. 2B) and a simultaneous decrease in multimeric mitochondrial cardiolipin (Fig. 2C), indicative of loss of cardiolipin from the inner mitochondrial leaflet. Finally, we noted significant morphological changes in the cells at higher levels (500 μM and 1 mM) of H2O2 exposure, with more cell rounding and fewer adherent cells, which is typically indicative of severe cell stress and/or death (Fig. 2).
On the basis of our finding that mitochondrial function was affected by H2O2 exposure, we next tested whether H2O2 exposure was associated with an increase in cellular O2•− production. We used EPR assay with the spin-trap compound CMH, which has been shown to specifically trap O2•− (13, 24), as well as DHE oxidation, to estimate cellular O2•− levels. For the EPR studies, to ensure that we estimated only O2•−, we added PEG-SOD (100 U/ml) before H2O2 exposure and found that the increase in the magnitude of the waveform as seen with H2O2 alone could be significantly reduced (Fig. 3A). Data from the EPR analyses showed significant increases in PEG-SOD-inhibitable cellular O2•− levels (Fig. 3, A and B). Significant increases were also seen in DHE oxidation after 4 h of exposure to H2O2 (Fig. 3C). Furthermore, utilizing MitoSOX Red to quantify alterations in mitochondrial O2•− production, we localized the increased O2•− to the mitochondrion (Fig. 3D).
In the context of recent reports that endogenous Zn2+ can affect cellular mitochondrial function (32, 49) and that Zn2+ levels rise in endothelial cells in response to oxidative stress (53), we hypothesized a potential role for intracellular free Zn2+ in the response following acute exposure to H2O2. To investigate this possibility, we used a Zn2+-specific fluorophore, FluoZin 3, to estimate Zn2+ levels after 4 h of exposure to varying concentrations of H2O2. By first incubating the cells and then performing the assay in serum-free medium, we eliminated any exogenous sources of Zn2+ and, thus, the need for external chelation. Our results identified significant increases in Zn2+ subsequent to H2O2 exposure (Fig. 4). Fluorescence microscopy using a second Zn2+ fluorophore, zinquin ethyl ester, yielded similar results (data not shown).
To better characterize the temporal nature of the cellular response to H2O2, we repeated certain experiments using a single concentration of H2O2, in this case 500 μM, over multiple time points (Fig. 5). Interestingly, on exposure to H2O2, we observed an almost immediate increase in intracellular Zn2+ levels (Fig. 5A). We also observed significant increases in MitoSOX fluorescence and LDH release ∼1 h after onset of exposure (Fig. 5, B and C). Between ∼2 and 4 h of exposure, we observed a large loss of Ψm as well as activation of caspases (Fig. 5, D and E). Finally, between ∼12 and 18 h, we observed a large increase in the percentage of TUNEL-positive cells (Fig. 5F).
Because H2O2 exposure resulted in increased levels of free Zn2+, we wished to determine whether this disruption in Zn2+ homeostasis was involved in the apoptotic cell death associated with acute exposure to H2O2. We overexpressed the Zn2+-sequestering protein (MT-1) or utilized the transition metal chelator TPEN. Through adenoviral transduction, we were able to significantly overexpress MT-1 protein (Fig. 6A). After transduction of MT-1, green fluorescent protein, or LacZ or exposure to TPEN, we explored the effect on cellular free Zn2+ and O2•− levels, mitochondrial function, and apoptotic cell death in response to acute H2O2 exposure. Overexpression of MT-1 or administration of 6.5 μM TPEN to PAECs significantly reduced H2O2-induced Zn2+ release (Fig. 6B) and increased O2•− levels (Fig. 7). Indeed, the reduction in O2•− was equivalent to that observed in PAECs treated with PEG-SOD (Fig. 7A vs. Fig. 3, B and C, respectively); however, the baseline EPR amplitude in MT-1-transduced cells was not significantly reduced, in contrast to that of PEG-SOD-treated cells (Fig. 7A), suggesting that MT-1 is not acting as an antioxidant. Additionally, overexpression of MT-1 or chelation of Zn2+ with TPEN resulted in decreased mitochondrial production (Fig. 7B), although TPEN was less effective at 1 mM H2O2. Overexpression of MT-1 also improved cytotoxic resistance to H2O2 as measured by LDH release (Fig. 8A). Interestingly, TPEN resulted in substantially greater cytotoxic resistance than MT-1 protein overexpression in cells exposed to 1 mM H2O2. In addition, we found that overexpression or chelation with TPEN decreased the adverse effects of H2O2 on Ψm (Fig. 8B). However, the specific mitochondrial protection of TPEN was not as effective as that of MT-1. Furthermore, MT-1 overexpression or TPEN chelation significantly reduced caspase activation (Fig. 8C) and TUNEL-positive cells (Fig. 8D), although TPEN significantly induced apoptosis in PAECs that were not exposed to H2O2. Overexpression of MT-1 caused a greater reduction of caspase activation and TUNEL-positive nuclei (Fig. 8, C and D) than LDH release (Fig. 8A), suggesting that cell death independent of apoptosis is also induced by H2O2 and may be less sensitive to alterations in cellular free Zn2+ levels.
In this study, we demonstrate a potentially important role for alterations in intracellular Zn2+ homeostasis in the cytotoxic and apoptotic effects associated with acute oxidative stress in PAECs. Our findings confirm similar studies in vitro (55) and in vivo models (39, 50) and extend previous findings in endothelial cells (41, 42). Our in vitro findings reported here also extend our recent in vivo data. We reported elevated levels of H2O2 in the pulmonary artery of lambs with ductal ligations designed to model persistent pulmonary hypertension in the newborn (58), a condition characterized by failure of pulmonary arteries to dilate at birth. Specifically, we report that increases in intracellular Zn2+ arising from strictly endogenous sources, after exposure to H2O2, occur with concomitant loss of mitochondrial function, as demonstrated by loss of Ψm, increases in mitochondrial O2•−, loss of cardiolipin from the inner mitochondrial leaflet, and induction of events associated with the apoptotic pathway. Significantly, when we overexpressed the MT-1 protein or utilized chelation therapy with TPEN, we observed a significant mitigating effect on Zn2+ release (Fig. 6B), in addition to protection from H2O2-mediated mitochondrial disruption (Fig. 8B) and from induction of apoptosis (Fig. 8, C and D). Although we observed significant protection from H2O2-mediated mitochondrial disruption and induction of apoptosis, the effect on necrotic cell death was only slightly improved (Fig. 8A), suggesting that at least two distinct pathways leading to cell death are active after acute bolus exposure to H2O2.
Taken together, our data indicate that if the cellular ability to regulate free Zn2+ within the cell is overwhelmed and/or the ability to detoxify ROS is impaired, a vicious cycle of cellular destruction occurs, forcing the cell into apoptosis or, in severe toxicity, necrotic cell death before the cell has the ability to complete the apoptotic process. Metal chelation with TPEN or overexpression of MT-1 protein in PAECs significantly reduced the induction of apoptosis (Fig. 8, B and C), yet the reduction in overall toxicity as measured by LDH release (Fig. 8A) was somewhat less. This result suggests that necrotic cell death can still occur and may not involve, or is less sensitive to, the cellular levels of free Zn2+. We speculate that these necrotic events may be due to direct oxidative stress-related effects of H2O2 on cellular macromolecules or H2O2 generation of ·OH through Fenton-type reactions with free ferrous iron that are likely not effected by Zn2+ sequestration. Interestingly, although metal chelation with TPEN caused a cell response similar to that caused by MT-1 protein overexpression, there were some notable differences. First, TPEN appears to sequester Zn2+ as efficiently as MT-1 protein (Fig. 6A). Furthermore, limitation of overall cellular O2•− production (Fig. 7B), but not suppression of mitochondrial-specific O2•− levels (Fig. 7B), was more effectively accomplished by metal chelation with TPEN than with MT-1 overexpression. In terms of overall cytotoxic protection, TPEN more effectively protected cells from acute necrotic cell death as measured by LDH release (Fig. 8A). Although it is considered to be preferential to Zn2+, TPEN is also capable of binding multiple divalent metal cations (2). Thus we attribute the differences in protection against necrotic cell death to the ability of TPEN to chelate other transition metals. For example, iron homeostasis could also be disrupted by acute rises in cellular H2O2, which could lead to the generation of ·OH through Fenton-type reactions and necrotic cell death secondary to disruption of the cell membrane due to the generation of lipid peroxides. This type of oxidative stress could be reduced by the chelation ability of TPEN, but not by MT-1. However, further studies are required to examine this possibility.
Furthermore, our data illustrate how MT-1 can function in intracellular binding and sequestration of Zn2+ and also the requirement for Zn2+ in the activation of apoptosis, but not necrosis. This assertion is supported by the early rapid rise of LDH release and O2•− production (Fig. 5, B and C), which we observed in <2 h, compared with loss of Ψm (Fig. 5D), caspase activation (Fig. 5E), and increases in percentage of TUNEL-positive cells (Fig. 5F); substantial changes required more time. Therefore, we believe that a large percentage of cells undergo necrosis within 2–4 h of acute exposure but that the cells that survive this initial pronecrotic insult proceed to cell death through a Zn2+-dependent apoptotic pathway (Fig. 9). Additionally, there is a discrepancy between reported physiological levels of H2O2 and the levels that can be reached during oxidative stress (for review see Refs. 10 and 20) as well as exposure as a consequence of a pathological condition (7, 37). Therefore, we utilized exposure levels that increased cellular levels of H2O2 in our PAECs by up to 50-fold over baseline levels but also were within the upper end of H2O2 levels reported in vivo.
MTs are small (∼6 kDa) metal-binding proteins (29). Although precise physiological roles for members of the MT family have yet to be defined, the hypothesized functions of MT-1 and MT-2 include detoxification of heavy metals, regulation of Zn2+ and copper homeostasis, and prevention of oxidative damage (40). Data indicate that MT may act as a cellular redox sensor and can regulate metal ion homeostasis through multiple cysteine residues forming metal thiolate clusters, allowing MT to potentially activate intracellular signal transduction pathways (25). On the basis of the findings reported here, we cannot rule out other metals known to bind to MT-1. The MT-1 protein can also act as a direct antioxidant or as a cofactor in a more general antioxidant response. However, our data argue against MT-1 acting as an antioxidant in this setting, because we did not see any significant difference in ROS levels between LacZ- or MT-1-transduced or TPEN-treated cells and the decreases in PEG-SOD-treated cells (Fig. 7A vs. Fig. 6, A and B). Therefore, we believe that these data provide evidence of a significant role for Zn2+ in mitochondrial disruption and apoptosis following acute oxidative stress.
A significant number of physiological, nutritional, and biochemical functions have been attributed to Zn2+ (4). Among those, numerous reports have concluded that Zn2+ functions as a cytoprotective agent, defending cells against oxidative insult and induction of apoptosis. There is evidence that Zn2+ requirements of the vascular endothelium are increased during inflammatory conditions such as atherosclerosis, where apoptotic cell death is prevalent (23). Proposed explanations for this protective effect include a role in cell membrane maintenance (38), activation of antiapoptotic signaling, blocking of proapoptotic NF-κB and AP-1 signal transduction (36), and inhibition of proapoptotic enzymatic activities, including caspase and endonuclease activity (8, 34), protein tyrosine phosphatase activity (19), and lipid peroxidation (6). However, it has also been shown that cells have a very tight regulatory apparatus in place for Zn2+, inasmuch as intracellular concentrations of free Zn2+ have been measured in pico- or nanomolar concentrations (12, 31). A loss of this Zn2+ homeostatic control causes cell death via apoptosis or, in extreme circumstances, necrosis. There is also in vivo evidence that excessive dietary Zn2+ intake can induce pathological conditions that have been associated with oxidative stress (61). Thus, in the same manner as calcium, damage from loss of intracellular Zn2+ regulation has been proven to be equal to damage from Zn2+ deficiency.
In terms of direct oxidative effects, it has been demonstrated that H2O2 and other ROS have the ability to oxidize proteins, which results in a loss of Zn2+ from the metal-binding domains of Zn2+-binding proteins (15, 51). Furthermore, it has been proposed that Zn2+, in high enough concentrations, has the ability to competitively displace copper and iron from heme domains, as well as the metal-binding domains of other proteins (22, 43, 45). Subsequently, copper and iron ions are highly redox reactive, and their release into the intracellular space allows rapid catalytic conversion of oxygen, in the presence of reducing agents, to potentially damaging ROS. These free radicals can include O2•− and ·OH (33, 43) and would potentially exacerbate the oxidative stress initially caused by excess H2O2.
We observed an increase in O2•− arising from the mitochondria in response to H2O2 exposure. During normal cellular respiration, mitochondrial-derived O2•− is spontaneously dismutated to H2O2 or is catalyzed by mitochondrial manganese SOD (SOD2). In this study, we observed an increase in mitochondrial O2•−, as well as loss of Ψm and cardiolipin, from the inner mitochondrial leaflet, suggesting that the induction of apoptosis could be arising through loss of cytochrome c. Although we cannot rule out cytosolic or membrane-bound producers of O2•−, such as NADPH oxidase, as additional sources of increased O2•−, generation from within the mitochondria has been directly implicated as an initiating event in the apoptotic process. Furthermore, our ability to significantly reduce the observed mitochondrial effects suggests a specific role for Zn2+ in mitochondrial dysfunction and induction of apoptotic cell death.
Because one primary function of pulmonary circulation is acquisition of oxygen for systemic needs, it stands to reason that an exquisite sensitivity of pulmonary endothelial cells to oxygen and ROS exists. ROS and, specifically, H2O2 have been proposed as signaling mediators in maintenance of adequate gas exchange in the lung by diverting blood flow from areas with low oxygen tension. Furthermore, production of ROS in pulmonary systems serves as a functional signal for leukocyte infiltration and induction of tissue-defense and wound-healing processes. Therefore, a severe acute upregulation of ROS and other signals of oxidative stress in the pulmonary system, including release of Zn2+, could result in inappropriate physiological responses, causing additional systemic oxidative stress and tissue damage. Thus, in clinical situations of acute oxidative stress, we hypothesize that proper management of intracellular Zn2+ balance must be considered to maximize the potential clinical benefits of Zn2+ and minimize potential Zn2+-mediated adverse consequences. Although true clinical improvements from antioxidant therapy remain elusive (27), our data suggest that timely and appropriate management of the multiple factors contributing to oxidative stress, including Zn2+ homeostasis, could potentially improve patient outcomes. However, this recommendation should be approached with caution, inasmuch as creation of a high concentration of MT or other free Zn2+ reduction factors could potentially result in Zn2+ deficiency (31). Therefore, we suspect that a critical balance between Zn2+ and Zn2+-sequestering proteins must be achieved to avoid adverse physiological consequences. Recent in vivo data are beginning to validate this postulation (28).
In conclusion, our data contribute to evidence of a biochemical link between increased H2O2 generation and alterations in cellular Zn2+ homeostasis that underlies the induction of apoptosis in PAECs. Our data also suggest that Zn2+ scavenging could be a potential therapy to reduce or prevent cardiovascular or other diseases associated with oxidative stress. However, the effect of this type of treatment in vivo is unknown, and direct assessment of such treatment requires animal studies.
This research was supported in part by National Institutes of Health Grants HL-60190, HD-39110, HL-67841, HL-72123, and HL-70061, American Heart Association Pacific Mountain Affiliates Grant 0550133Z, and a LeDucq Foundation Award (all to S. M. Black). D. A. Wiseman was supported in part by National Heart, Lung, and Blood Institute Grant T32 HL-66993.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society