| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Nelson Institute of Environmental Medicine, New York University School of Medicine, Tuxedo, New York 10987
1To whom requests for reprints should be addressed at Nelson Institute of Environmental Medicine, New York University School of Medicine, 57 Old Forge Road, Tuxedo, NY 10987. E-mail: costam{at}env.med.nyu.edu or costam01{at}nyu.edu
Abstract
Iron is an essential nutrient to most organisms, and is actively involved in oxygen delivery, electron transport, DNA synthesis, and many other biochemical reactions important for cell survival. We previously reported that nickel (Ni) ion exposure decreases cellular iron level and converts cytosolic aconitase (c-aconitase) to iron-regulatory protein-1 in A549 cells (Chen H, Davidson T, Singleton S, Garrick MD, Costa M. Toxicol Appl Pharmacol 206:275–287, 2005). Here, we further investigated the effect of Ni ion exposure on the activity of mitochondrial iron-sulfur (Fe-S) enzymes and cellular energy metabolism. We found that acute Ni ion treatment up to 1 mM exhibits minimal toxicity in A549 cells. Ni ion treatment decreases the activity of several Fe-S enzymes related to cellular energy metabolism, including mitochondrial aconitase (m-aconitase), succinate dehydrogen-ase (SDH), and NADH:ubiquinone oxidoreductase (complex I). Low doses of Ni ion for 4 weeks resulted in an increased cellular glycolysis and NADH to NAD+ (NADH/NAD+) ratio, although glycolysis was inhibited at higher levels. Collectively, our results show that Ni ions decrease the activity of cellular iron (Fe)-containing enzymes, inhibit oxidative phosphorylation (OxPhos), and increase cellular glycolytic activity. Since increased glycolysis is one of the fundamental alterations of energy metabolism in cancer cells (the Warburg effect), the inhibition of Fe-S enzymes and subsequent changes in cellular energy metabolism caused by Ni ions may play an important role in Ni carcinogenesis.
Key Words: Fe-S enzymes • NADH/NAD+ ratio • glycolysis • oxidative phosphorylation
Introduction
Iron is an essential nutrient to most organisms, and is actively involved in oxygen delivery, electron transport, DNA synthesis, and many other biochemical reactions important for cell survival (1). However, excess free-available iron has the potential to harm the cell by generating reactive oxygen species through the Fenton reaction (2). To maintain an adequate iron level, the cell has developed sophisticated regulatory mechanisms, namely iron regulatory proteins (IRPs), to control the uptake, utilization, and storage of iron. IRP-1 and IRP-2 sense the level of free-available intracellular iron, and post-transcriptionally modulate the expression of genes that encode transferrin receptor (TfR) and ferritin by binding to the iron response elements (IREs) present in their mRNAs (3). When activated under an iron-depleted condition, IRPs stabilize TfR mRNA and block the synthesis of ferritin, and thus, favor iron uptake over storage. Conversely, when the iron level is high, downregulation of IRP activity results in enhanced translation of ferritin mRNA and decreased stability of TfR mRNA. IRP-2 is mainly degraded by the proteasome during iron-repletion, while IRP-1 is inactivated by receiving a [4Fe-4S] cluster and is converted to cytosolic aconitase (c-aconitase), an enzyme which has the same activity as its counterpart, mitochondrial aconitase (m-aconitase).
Glycolysis, the tricarboxylic acid (TCA) cycle, and oxidative phosphorylation (OxPhos) are central biochemical processes in cellular energy metabolism, and are linked to cellular iron homeostasis. For instance, the expression of several genes that encode glycolytic enzymes is upregulated by the transcription factor hypoxia inducible factor-1 alpha (HIF-1
), which is stabilized and transactivated by the inhibition of several Fe(II)-2-oxoglutarate–dependent dioxygenases under iron-deficient conditions (4). Moreover, the availability of cellular iron may affect the TCA cycle and OxPhos by modulating the gene expression and/or activity of several iron-sulfur (Fe-S) enzymes, including m-aconi-tase, succinate dehydrogenase (SDH), and NADH:ubiqui-none oxidoreductase (complex I). Iron deficiency can not only directly inhibit the biosynthesis of Fe-S clusters, but it can also block m-aconitase protein synthesis in mammalian cells by causing IRPs to bind to the IRE in its mRNA 5' untranslated region (5'-UTR). Interestingly, IREs or IRE-like elements have also been found in mRNAs that encode other Fe-S proteins in some eukaryotes. In Drosophila melanogaster, the 5'-UTR of SDH subunit b mRNA has an IRE that mediates a similar response to iron deficiency as m-aconitase (5). Recently, it was reported that synthesis of the 75-kDa subunit of mitochondrial complex I may also be regulated by an IRE-like stem-loop structure in the 5'-UTR of its mRNA (6).
We previously reported that nickel (Ni) ion exposure for 24 hrs decreased cellular iron by about 40% and converted c-aconitase into IRP-1 in lung carcinoma A549 cells (7). These results prompted us to test the hypothesis that Ni ion exposure decreases the activity of other cellular iron-dependent enzymes, such as those with Fe-S clusters in the TCA cycle and OxPhos. We found that Ni ion exposure decreases m-aconitase, SDH and mitochondrial complex I activities, but increases the activities of two other TCA cycle enzymes that do not rely on iron for their function. The possible impact of Ni ion exposure on cellular energy metabolism was also assessed and is discussed in this manuscript.
Materials and Methods
Chemicals.
All reagents were obtained from Sigma (St. Louis, MO) unless otherwise specified.
Cell Culture.
Human lung carcinoma A549 cells were maintained in Ham’s F-12 K medium (GIBCO BRL, Grand Island, NY) supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. Cells were passaged two to three times each week and kept at 37°C in a humidified 5% CO2 atmosphere.
Colony Formation Assay.
After trypsinization, 1 x 106 cells were seeded into each dish (100-mm diameter). Cells were allowed to attach overnight, and then exposed to different concentrations of nickel chloride (NiCl2) for 24 hrs. Cells were rinsed twice with prewarmed medium and then trypsinized. After neutralizing with complete medium, the cell suspension was passed through a 40 µm Falcon nylon cell strainer (BD Biosciences, San Jose, CA) to eliminate cell clumps. Two hundred cells were then reseeded into each of three dishes (150-mm diameter), and grown for 2 weeks. Surviving colonies were stained with Giemsa stain and counted. All experiments were conducted in triplicate.
Aconitase and Glutathione Peroxidase Activity.
Cytosolic and mitochondrial aconitase activities were measured as previously described (7). Cellular selenium-dependent glutathione peroxidase activity was measured using a Glutathione Peroxidase Cellular Activity Assay kit (Sigma). All experiments were conducted in triplicate.
Mitochondria Isolation.
About 2 x 107 cells were collected and the cell pellet was resuspended in M buffer (250 mM sucrose, 2 mM Hepes, pH 7.4, 0.1 mM ethylene glycol bis[2-aminoethyl ether]-N,N,N'N’-tetraacetic acid [EGTA]). After lysis with a glass-Teflon homogenizer by upward and downward motion 10 times, the homogenate was centrifuged at 600 g for 6 min. The supernatant was transferred to a clean Eppendorf tube, and the pellet was resuspended, homogenized, and centrifuged again. The supernatant collected from two homogenizations was combined and centrifuged at 10,000 g for 10 min. The pellet was washed once and resuspended in the appropriate buffer for the following mitochondrial enzyme assays.
Isocitrate Dehydrogenase (IDH) and Malate Dehydrogenase (MDH) Activity.
IDH and MDH activities were measured as previously described (8). In brief, mitochondria were resuspended in PBS with 0.1% Triton X-100, and sonicated twice by applying 10 x 1-sec pulse with a Branson Sonifier 450 (Branson Ultrasonics Co., Danbury, CT). After centrifugation at 14,000 g for 30 min, the supernatant was transferred to a clean Eppendorf tube. After measuring protein concentration using the Bio-Rad DC protein assay (Bio-Rad, Richmond, CA), 50 µg supernatant was used to measure IDH and MDH activity. For the IDH assay, each assay contained 100 mM Tris, pH 7.4; 2 mM NADP+; 2 mM magnesium chloride (MgCl2); and 5 mM isocitrate in a final volume of 1 ml. For the MDH assay, each assay contained 100 mM Hepes-Tris, pH 7.4; 0.16 mM NADH; and 0.13 mM oxaloacetate in a final volume of 1 ml. Enzyme activity was measured by following the linear absorbance change at 340 nm at 30°C for 30 min using a HTS 7000 Plate Reader (Perkin Elmer, Wellesley, MA). An extinction coefficient of 6.81 mM–1 was used for NADH. All experiments were conducted in triplicate.
Complex I (NADH:ubiquinone oxidoreductase) and II (SDH) Activity.
Complex I activity was measured as described by Ben-Shachar et al. (9). Briefly, mitochondria were resuspended in hypotonic buffer (25 mM potassium phosphate, pH 7.4 and 5 mM MgCl2), and freeze-thawed three times. The sample was subsequently sonicated by applying a 10 x 1-sec pulse with a Branson Sonifier 450. Eighty micrograms of mitochondrial suspension were used to measure complex I activity in 20 mM potassium phosphate, pH 7.2; 5 mM MgCl2; 1 mM potassium cyanide (KCN); 0.14 mM NADH; and 50 µM Coenzyme Q1 (Sigma) in the presence or absence of 20 µM rotenone. Oxidation of NADH was followed at 340 nm for 5 min. Complex I activity is reported as the difference between the rate of NADH oxidation with and without rotenone. SDH activity was measured as described by Parker et al. (10). In brief, the pellet was resuspended in hypotonic buffer (25 mM potassium phosphate, pH 7.4 and 5 mM MgCl2) and snap-frozen in liquid nitrogen. After quickly thawing in a room temperature water bath, samples were promptly chilled to 4°C. Thirty micrograms of mitochondrial suspension were used to measure SDH activity by following the succinate-dependent reduction of 2,6-dichlorophenol-indophenol (DCIP) at 590 nm with or without 10 mM malonate. An extinction coefficient of 19.1 mM–1 was used for DCIP. SDH activity is reported as the difference between the rate of DCIP reduction with and without malonate. All experiments were conducted in triplicate.
Glucose and Lactate Measurement.
The concentration of glucose in the medium was measured using a Glucose Assay kit (Sigma). The concentration of lactate in the medium was measured as previously described (11). To measure lactate concentration, 0.5 ml cell medium was collected in an Eppendorf tube and deproteinized with 315 µl 7% perchloric acid. After incubation on ice for 15 min, samples were centrifuged at 14,000 g for 5 min and the supernatant was collected and adjusted to pH 6.0 with 1 M potassium hydroxide (KOH). Samples were centrifuged to remove potassium perchlorate. Twenty microliters of supernatant were put into a 24-well plate and incubated with 80 µmol glycine buffer, pH 10.0, 10 units lactate dehydrogenase (Sigma), and 2 mg NAD+ in a final volume of 1.0 ml. After incubation at 25°C for 1 hr, the concentration of lactate was measured as the change in O.D. at 340 nm. Lactate standard solutions were assayed simultaneously. All experiments were conducted in triplicate.
NADH and NAD+ Measurement.
The concentrations of cellular NADH and NAD+ were measured as previously described (12). Cells were washed twice with ice cold PBS and lysed in 1 ml ice-cold extraction buffer (10 mM nicotinamide, 20 mM sodium bicarbonate [NaHCO3], and 100 mM bisodium carbonate [Na2CO3]). The scraped lysate was frozen in a dry ice–ethanol bath, thawed quickly in a room temperature water bath, and then kept on ice. The samples were centrifuged once to eliminate insoluble material. The lysate was then diluted to 200 µg/ml. Twenty-microgram samples, in triplicate, were directly added into a 96-well plate, and total NAD (NADH + NAD+) concentration was measured using a cycling assay as described by Jacobson et al. (13). Each assay contained 100 mM bicine, pH 7.8, 500 mM ethanol, 0.42 mM 3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide (MTT) tetrazolium, 1.66 mM phenazine ethosulfate, 4.16 mM EDTA, 0.83 mg/ml bovine serum albumin, and 14 units alcohol dehydrogenase in a final volume of 200 µl. The amount of total NAD was measured as the O.D. change at 590 nm at 37°C for 5 min with a HTS 7000 Plate Reader. For NADH measurement, the lysate was incubated at 60°C for 30 min to destroy NAD+, and promptly chilled to 4°C. Twenty-microgram samples, in triplicate, were used to measure NADH concentration with the same cycling assay used to measure NAD concentration. NADH standard solutions were assayed simultaneously. The ratio of NADH/NAD+ was calculated based on the results of total NAD and NADH concentrations. All experiments were conducted in triplicate.
ATP Measurement.
Intracellular ATP was measured using an ENLITEN ATP Assay System kit (Promega, Madison, WI). Cells were rinsed twice with ice cold PBS and intracellular ATP was extracted in 0.5% trichloroacetic acid. After centrifugation at 20,000 g for 5 min, the supernatant was collected for ATP measurement and the pellet was dissolved in 0.5 N sodium hydroxide (NaOH) for protein measurement. Prior to ATP measurement, the pH of the supernatant was adjusted to pH 7.75 by adding 40 volumes of 25 mM Tricine buffer (pH 7.8). The light intensity emitted after adding rL/L reagent (Promega) was determined using a luminometer (Wallac 1420 Victor multilabel counter system, Perkin Elmer), with a delay of 2 sec and integration time of 10 sec. All experiments were conducted in triplicate.
Statistical Analysis.
Two-tailed Student’s t tests were used to determine the significance of differences in enzyme activity or metabolite concentration between treated samples and controls. Differences were considered significant at a P < 0.05.
Results
Toxicity of Ni Ion Exposure in A549 Cells.
In contrast to many other transition metal ions, mammalian cells tolerate a high level of soluble nickel. To determine Ni ion toxicity in lung carcinoma A549 cells, we measured their clonogenic survival following Ni ion exposure. After treating cells with various concentrations of NiCl2 for 24 hrs, equal numbers of A549 cells were reseeded and grown in complete medium in the absence of Ni ions. The number of colonies was counted 2 weeks later. As shown in Figure 1
, the colonogenic rate of A549 cells was not affected by 1 mM Ni ions, but decreased by 20% at 2 mM Ni ions, and decreased more strikingly at 5 mM Ni ions. These data are consistent with our previous MTT viability studies, which showed that 1 mM Ni ions did not cause a significant loss of A549 cell viability (14). Based on these results, we exposed cells to a maximum of 1 mM Ni ions in this study.
|
). Meanwhile, the activities of several Fe-containing enzymes, including m-aconitase, SDH, and complex I, decreased by 40%–50% compared with controls (Table 1). The same Ni ion exposure increased the activities of two Fe-lacking enzymes in the TCA cycle: IDH and MDH (Table 1). Furthermore, Ni ion treatment did not affect the activity of selenium-dependent glutathione peroxidase (Table 1).
|
). KCN prevents NADH oxidation by inhibiting mitochondrial electron transport at complex IV, and we observed that it caused an increase of glycolytic activity to maintain intracellular ATP level (Fig. 2a–d). Iodoacetate (IAA) is an inhibitor of the glycolytic enzyme glyceraldehyde 3-phosphate dehydrogenase; it decreased the NADH/NAD+ ratio, glycolytic activity, and intracellular ATP level (Fig. 2a–d).
|
, iron chelation with DFX increased the NADH/NAD+ ratio, glycolytic activity, and intracellular ATP level, which indicates that cellular OxPhos was blocked. Although the NADH/NAD+ ratio was also elevated by 500 µM NiCl2 treatment, a subsequent increase in glycolytic activity was not observed (Fig. 3a–d). Increasing the Ni ion concentration to 1 mM did not increase the NADH/NAD+ratio and glycolytic activity (Fig. 3a–d). Instead, it decreased the consumption rate of glucose from the medium (Fig. 3a), indicating that cellular glycolysis was inhibited by the high concentration of Ni ions. Since inhibition of glycolysis became more prominent in the presence of 1 mM Ni ions, the NADH/NAD+ ratio decreased, despite the fact that it was elevated in 0.5 mM Ni ion concentration (Fig. 3a, d).
|
). These results indicate that Ni ion exposure impairs mitochondrial OxPhos, and forces cells to increase glycolytic activity to meet their energy needs.
|
Nickel compounds are harmful to both human health and our living environment (15). Several lines of evidence suggest that nickel compounds interfere with cellular iron metabolism. For instance, Ni ions interfere with the functions of several Fe(II)-dependent dioxygenases and cause a hypoxic condition in cells under normal oxygen tension by stabilizing and activating HIF-1 (16); Ni ions upregulate several genes related to cellular iron metabolism, including TfR and heme oxygenase-1 (17); and iron inhibits nickel subsulfide (Ni3S2) carcinogenesis in rat skeletal muscle (18). In our previous study, we reported that Ni ion exposure decreases the level of cellular iron and activates IRP-1 by competing for iron uptake at the divalent metal transporter-1 (7). Here, we further investigated the effect of Ni ion exposure on the functions of several Fe-containing enzymes involved in cellular energy metabolism. Our results demonstrate that Ni ion exposure decreases the activity of these Fe-containing enzymes, and increases the activity of two Fe-lacking enzymes in the TCA cycle. Long-term exposure to Ni ions leads to an increased cellular glycolysis rate and intracellular NADH/NAD+ ratio, although a higher concentration of Ni ions inhibits glycolysis.
Most Fe-containing enzymes involved in cellular energy metabolism contain Fe-S clusters, which are non-heme prosthetic groups and important electron carriers. Besides its inhibition of c- and m-aconitase activities as shown in our previous study (7), Ni ion exposure also decreases the activities of other enzymes with a similar Fe-S cluster, such as succinate aconitase and complex I. These findings are consistent with previous reports that iron deficiency decreases m-aconitase, succinate aconitase, and complex I activities (20–23). The decrease of m-aconitase, SDH, and complex I activity by Ni ions is likely to result from translation inhibition caused by increased IRE-IRP interactions, as well as destabilization of Fe-S clusters in the enzymes’ active centers by iron deficiency. In comparison, Ni ions increase the activities of two Fe-lacking enzymes, MDH and IDH. Since IDH is considered one of the rate-limiting enzymes in the TCA cycle, an increase of its activity could reflect the cell’s effort to restore TCA cycle function following Ni ion exposure. This notion is supported by previous reports that several Fe-lacking mitochondrial enzymes, including IDH and MDH, become activated in rat muscle tissue after an iron-deficient diet (20, 21).
Cellular OxPhos relies on iron for the transfer of an electron from NADH to oxygen. Disruption of OxPhos by iron depletion leads to an accumulation of NADH and an increase in the rate of glycolysis, as was evidenced by DFX treatment. It is still unknown how DFX increases the intracellular ATP level, although a similar finding was also reported by another group (24). It was surprising that a high concentration of Ni ions (0.5–1 mM) did not produce changes in cellular energy metabolism similar to DFX. The dose-dependent decrease in the glucose consumption rate indicates that glycolysis is inhibited by a high concentration of Ni ions. After decreasing the Ni ion concentration and prolonging the exposure time, the cells clearly demonstrated increased glycolytic activity and an increased NADH/NAD+ ratio—a pattern of changes similar to those seen during iron-deficiency. Since decreased OxPhos and increased glycolysis are among the fundamental changes undergone by cancer cells (the Warburg effect) and are often associated with an aggressive phenotype and resistance to therapeutic agents (19), the inhibition of OxPhos and activation of glycolysis by Ni ions may play an important role in nickel carcinogenesis.
Acknowledgments
We thank Jingxia Li for her assistance with ATP measurements and Juliana Powell for her excellent secretarial support.
Footnotes
This work was supported by grant ES00260, ES10344, and T32-ES07324 from the National Institute of Environmental Health Sciences, and CA16087 from the National Cancer Institute.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
