ZINC TOXICITY ON ATP PRODUCTION175 Copyright © 2007 John Wiley & Sons, Ltd.J. Appl. Toxicol. 2008; 28: 175–182 DOI: 10.1002/jatJOURNAL OF APPLIED TOXICOLOGYJ. Appl. Toxicol. 2008; 28: 175–182 Published online 20 June 2007 in Wiley InterScience(www.interscience.wiley.com) DOI: 10.1002/jat.1263
Zinc toxicity alters mitochondrial metabolism
and leads to decreased ATP production in hepatocytes
Joseph Lemire, Ryan Mailloux and Vasu D. Appanna*
Department of Chemistry and Biochemistry, Laurentian University, Sudbury, Ontario P3E 2C6, Canada
Received 9 February 2007; Revised 19 March 2007; Accepted 30 March 2007ABSTRACT: Although zinc (Zn) is a known environmental toxicant, its impact on the cellular energy-producing machinery is not well established. This study investigated the influence of this divalent metal on the oxidative ATP produc- ing network in human hepatocellular carcinoma (HepG2) cells. Zn-challenged cells contained more oxidized proteins andlipids compared with control cells. Zn severely impeded mitochondrial functions by inhibiting aconitase, α α α α α-ketoglutarate dehydrogenase, isocitrate dehydrogenase-NAD+ + + + + dependent, succinate dehydrogenase and cytochrome C oxidase Zn-exposed cells had a disparate mitochondrial metabolism compared with the control cells and produced significantly less ATP.However, the expression of isocitrate dehydrogenase-NADP+ + + + + dependent was more prominent in cells treated with Zn. Hence, Zn-induced pathologies may be due to the inability of the mitochondria to generate energy effectively.Copyright © 2007 John Wiley & Sons, Ltd. KEY WORDS: zinc; mitochondrial metabolism; ATP production; aconitase; succinate dehydrogenase; TCA cycleof aluminum (Al) to inhibit the TCA cycle enzymes, to impede ATP synthesis and to promote anaerobiosis viaHIF-1α stabilization was recently demonstrated (Maillouxand Appanna, 2007; Mailloux et al., 2006a, 2006b). Zn has been shown to promote ROS (Reactive Oxygen Species) via the displacement of bioavailable Fe, thebinding of GSH (Glutathione) molecules, and it is knownto inhibit some key TCA cycle enzymes (Dinely et al., 2003; Gazaryan et al., 2002; Koh et al., 1996; Sensi et al., 1999). However, the exact mechanism(s) of how this divalent metal interferes with oxidative-ATP produc-tion is not fully understood. This report describes the ability of Zn to inhibit numerous mitochondrial enzymesand to perturb ATP production in human liver cells. Theimplication of the diminished activities of aconitase (ACN) and cytochrome C oxidase (Cyt C Ox) on various disease processes, as a consequence of Zn toxicity, isalso discussed.
Cell Culture, Isolation and Fractionation
Human hepatocellular carcinoma (Hep G2) cells werea gift from Dr Templeton (University of Toronto) and were maintained in α-MEM containing 5% FBS and 1% antibiotics. The cells were routinely seeded at100000cellsml−1 in 175cm2 culture flasks and incubatedwith 5% CO2 in a humidified atmosphere operating at 37°C. Upon reaching 70% confluency the cell mono-layer was washed with PBS and the cultures were
Zinc is an essential trace element due to its participationin a variety of biological processes. Zinc is involved in numerous enzymatic reactions, plays a pivotal role in transcription, cell signaling and in the regulation ofcellular pH (Hambidge, 2000). Zinc deficiency is often characterized by compromised T-cell function, neurosensory injury and hindered healing processes (Piao et al., 2003;Prasad, 1996; Sandstead, 1991). However, in elevated concentrations, Zn is toxic. It is a significant componentof ambient particulate matter (PM) and it is also an important pollutant in industrial settings (Barceloux, 1999; Coleman, 1992). This divalent metal has beenimplicated in neurological disorders, in the disruption ofiron homeostasis, and has also been shown to perturb cholesterol metabolism (Fosmire, 1990; Religa et al., 2006). Although there have been numerous studies on Zn toxicity, there is a dearth of information on the influenceof this metal on mitochondrial energy production. In aerobic organisms, the mitochondria are the mainsupplier of energy and are the site of the tricarboxylicacid (TCA) cycle, a metabolic network involved in the generation of reducing factors that power the productionof ATP (Costello et al., 1997; Fernie et al., 2004; Manevet al., 1997). Within this organelle reside numerousenzymes that may be potential targets of Zn. The ability * Correspondence to: V. D. Appanna, Department of Chemistry and Bio-chemistry, Laurentian University, Sudbury, Ontario P3E 2C6, Canada.E-mail: VAppanna@laurentian.caContract/grant sponsor: Industry Canada.
176J. LEMIRE ET AL.Copyright © 2007 John Wiley & Sons, Ltd. J. Appl. Toxicol. 2008; 28: 175–182DOI: 10.1002/jat re-supplemented with serum-free media reconstituted with2.5mM citrate (control), 2.5mM citrate: 50 μM ZnCl2 [note that the normal serum concentration of Zn is 15 μM (Steinebach and Wolterbeek, 1993; Takeda et al., 2001; Walsh et al., 1994)] or 2.5mM citrate: 50 μM H2O2. Theseconcentrations of Zn and H2O2 are known to be toxic(Abordo et al., 1999; Canzoniero et al., 1999; Dumontet al., 1999). Cell viability was monitored with trypanblue exclusion assay (Shannon, 1978). Following the 24h stress period, the media was removed and the cells werewashed with PBS [136.8mM sodium chloride, 2.5mM potassium chloride, 1.83mM dibasic sodium phosphate,and 0.431mM monobasic potassium phosphate (pH 7.4)]. Cells were harvested by trypsinization and then centri-fuged at 250g for 10min at 4°C. The cellular pellet was resuspended in cell storage buffer (CSB) (50mMTris-HCl, 1mM phenylmethylsulphonylfluoride, 1mM dithiothreitol, 250mM sucrose, 2mM citrate, containing1mgml−1 of pepstatin A and 0.1mgml−1 of leupeptin) and stored at −86°C until needed. The cells were thawedand pelleted by centrifugation at 250g for 10min at4°C. The pellet was re-suspended in a minimal volumeof CSB (4 × 106 cells/50 μl). The resultant cell suspen- sion was disrupted on ice using a Brunswick sonicator, operating for 5s with 1s bursts. Whole cells and nucleiwere removed by centrifugation at 850g for 10min at 4°C. The mitochondria were then isolated from thecytoplasm via differential centrifugation, 12000g for 30min at 4°C. Following the centrifugation, the solublefraction (supernatant) was placed in an ice cold eppendorfand the mitochondrial pellet was resuspended in aminimal amount of CSB. Protein quantification wasascertained via the Bradford assay (Bradford, 1976) andBSA was utilized as the standard to normalize the assay.
Oxidized Protein and Lipid Analyses
To evaluate oxidized lipid content in the membrane, thethiobarbituric acid reactive species assay (TBARS) wasperformed as described in Aydin et al. (2005). The frac- tion containing the membrane portion was solubilized ina mixture containing 15% trichloroacetic acid, 0.375%trichlorobarbituric acid and 0.25N HCl. The reactionmixture was subsequently heated for 20min at 100°C. The precipitated protein was then pelleted at 21000g for10min and the supernatant was monitored at 532nm. Reaction mixtures lacking trichlorobarbituric acid wereused as negative controls. Protein carbonyl content was determined by perform-ing a dinitrophenyl hydrazine (DNPH) assay as describedin Costello et al. (1997). One mg of soluble protein was homogenized with 1ml of 2% (w/v) DNPH and reactedfor 1h. The protein was subsequently precipitated and the pellet was washed thrice with ethylacetate:ethanol (1:1).1ml of 6M guanidine-HCl was added to the mixture andread spectrophotometrically at 360nm. Reaction mixtureslacking DNPH were used as negative controls.
Blue Native PAGE (BN PAGE) and Protein Activity
BN PAGE was performed according to the method described by Beriault and Schagger (Beriault et al., 2007;Schagger and von Jagow, 1991). 4–10% gradient gelswere cast in a BioRad MiniProtean™ 2 electrophoresis unit. Samples of 2 μg of protein equivalent μl−1 wereprepared in blue native buffer (500mM 6-amino hexanoicacid, 50mM BisTris (pH 7.0) and 1% β-dodecyl-D-maltoside). β-dodecyl-D-maltoside was omitted in the case of soluble proteins. 30 μg of the prepared protein samples was loaded into each well of the native gel. For migration through the stacking gel, the unit was run at80V. For the resolving gel the electrophoresis was run at200V. The blue cathode buffer [50mM Tricine, 15mM BisTris, 0.02% w/v Coomassie G-250 (pH 7.0) at 4°C]was exchanged for colourless cathode [50mM Tricine, 15mM BisTris (pH 7.0) at 4°C] once the running frontwas half way through the resolving gel. Upon comple-tion, the gel slab was removed and incubated in an equi- libration buffer (25mM Tris-HCl, 5mM MgCl2 (pH 7.4) for 15min. Enzyme activity was visualized with the aidof formazan precipitation. The gels were incubated inreaction mixture containing equilibration buffer, 5mM substrate, 0.5mM cofactor, 0.5mgml−1 iodonitrotetrazo- lium chloride (INT) and 0.2mgml−1 phenazine metho- sulphate (PMS). Isocitrate dehydrogenase (ICDH)activitywas ascertained with 5mM isocitrate, 0.1mM NAD+ orNADP+, PMS and INT. α-Ketoglutarate dehydrogenase(α-KGDH) activity was made apparent with 5mM α-ketoglutarate, 0.1mM NAD+, 0.25mM CoA, PMS andINT. The concentration of NAD+ and NADP+ used was0.1mM to prevent cross reactivity. Cyt C Ox activity wasdeduced with the utilization of equilibration buffer supplemented with KCN (5mM), 10mgml−1 of diamino- benzidine, 10mgml−1 cytochrome C and 562.5mgml−1 of sucrose.Two dimensional SDS-PAGE was performed on theactivity bands as described in (Mailloux et al., 2006b).Band specificity was confirmed by using standard en- zymes. The purity of mitochondrial and cytoplasmic frac-tions was confirmed by VDAC (voltage dependent anion channels) and F-Actin immunoblots. Activity bands weresubsequently quantified using Scion imaging software forWindows.
SDS PAGE and 2D SDS-PAGE gels were performedaccording to the modified method described by Laemmli
ZINC TOXICITY ON ATP PRODUCTION177 Copyright © 2007 John Wiley & Sons, Ltd.J. Appl. Toxicol. 2008; 28: 175–182DOI: 10.1002/jat (1970). Protein samples were first solubilized in 62.5mM Tris-HCl (pH 6.8), 2% SDS and 2% β-mercaptoethanol at 100°C for 5min. The protein samples were thenloadedin a 10% isocratic gel and electrophoresed using a dis- continuous buffer system. For 2D immunoblot analysis, activity bands from native gels were precision cut fromthe gel and incubated in denaturing buffer (1% β- mercaptoethanol, 5% SDS) for 30min, and then placed vertically into the well of the SDS gel. Following com- pletion of the SDS electrophoresis, the proteins were trans-ferred electrophoretically to a Hybond™ — Ppolyvinylidene difluoride membrane for immunoblotting. Non-specific binding sites were blocked by treating the membranewith 5% non fat skim milk dissolved in TTBS [20mMTris-HCl, 0.8% NaCl, 1% Tween-20 (pH 7.6)] for 1h.Polyclonal antibodies raised against mitochondrialNADP+-ICDH, ACN and KGDH were generous giftsfrom Dr S. Yokota, University of Yamanashi, Dr R. Eisenstein, University of Wisconsin-Madison and Dr G. Lindsay, University of Glasgow, respectively. The sec-ondary antibodies (Santa Cruz) consisted of horseradish peroxidase-conjugated mouse anti-rabbit. The detectionrelied on incubation of the probed membrane for 5min at room temperature in the presence of Chemiglow reagent (Alpha Innotech). Visualization of the immuno-blot was documented via a ChemiDoc XRS system(Biorad Imaging Systems).
To further ascertain the effect of zinc on the mito- chondria, mitochondrial isolate (2mgml−1 protein equiva-lent) obtained from control and Zn stressed conditionswere incubated in a phosphate buffer [10mM phosphate,5mM MgCl2 (pH 7.4)] containing 1mM citrate, 0.1mMNAD+ and 0.5mM ADP for 1h at 37°C. The reaction was stopped by the addition of ice cold 0.1% perchloricacid. The organic acids and nucleotides were extracted for HPLC analysis (Samizo et al., 2001). The resultant spent fluid from the reaction was analysed using a C18-reverse phase column (Phenomenex) with the aid of anAlliance HPLC (Waters). The mobile phase used con-sisted of 20mM KH2PO4 (pH 2.9 with 6N HCl),operatingat an elution rate of 0.7mlmin−1 at ambient temperature. The identities of the metabolites were compared with knownstandards, or reaction mixtures were spiked with the appropriate standards. The initial levels of metaboliteswere obtained by running the reaction mixture at time zero.
The impact of Zn on mitochondrial activity was achievedusing immunofluorescence. Cells were grown to a mini- mal density on coverslips and incubated for 24h in thepresence of serum-free α-MEM containing either 2.5mMcit or 2.5mM cit: 50 μM ZnCl2. The coverslips werewashed once with 0.5mM EDTA and twice with PBS to ensure all residual zinc was removed. The cells were incubated in 10 μgml−1 Rhodamine B, diluted in α- MEM for 20min at 37°C. The cells were then fixedwith methanol:acetic acid (3:1). Coverslips were thensubmerged in a Tween-20 Tris-buffered saline (TTBS) solution with 5% FBS for 1h to permeabilize the mem-brane. The coverslip was then washed three times withTris-buffered saline (TBS). The hepatocytes were then incubated for 2h in anti-VDAC antibody diluted (1/400)in TBS/5% FBS solution. The coverslips were washed thoroughly and incubated in anti-rabbit FITC conjugate.Following antibody exposure the cells were washed andmounted onto microscope slides. The cells were visual- ized using an inverted deconvolution microscope (Zeiss).The protonated Rhodamine B molecule was detected at
λexcitation = 564nm and λemission = 620nm. Similarly the fluorescein isothiocyanate was dectected at λexcitation=
495nm and λemission = 520nm, allowing the distinction ofthe mitochondria.
All experiments were done three times and in duplicate (where appropriate standard deviations are given). Statis-tical significance was analysed using Student’s t-test.
Results and Discussion
In an effort to evaluate the toxic impact of zinc on energy production, HepG2 cells were exposed to zinc for24h and mitochondrial functions were assessed. 50 μMZn has been shown to be associated with Zn toxicity in humans (Steinebach and Wolterbeek, 1993; Walsh et al.,1994). A 25% reduction in cell viability was observed.Rhodamine B revealed a diminished proton gradient inHepG2 mitochondria exposed to Zn (Fig. 1). The moreintense fluorescence was an indication of an increasedproton gradient in the control cells. Rhodamine B is known to provide a visual indication of the activity of the mitochondrion. The interaction between the Rhodamine Band the mitochondrion was further confirmed by thedetection of the voltage dependent mitochondrial porinVDAC, an important mitochondrial marker. The spotsdue to Rhodamine B fluorescence correspond to those ofthe VDAC (FITC) fluorescence, thus confirming thelocalization of the mitochondria. This decrease in mito- chondrial fluorescence in Zn-challenged cells promptedus to evaluate the oxidative stress triggered by Zn. Indeed, a drastic increase in oxidized lipids and proteins was observed in the Zn-stressed cells (Table 1). Thiswould undoubtedly lead to mitochondrial dysfunction. As
178J. LEMIRE ET AL.Copyright © 2007 John Wiley & Sons, Ltd.J. Appl. Toxicol. 2008; 28: 175–182DOI: 10.1002/jatZn seemed to be interfering with aerobic energy metabo- lism, the mitochondria’s capability to metabolize citratewas assessed. Mitochondria from control and Zn-stressedHepG2 cells that were incubated for 1h in the presenceof citrate and NAD+ were analysed by HPLC. Highercitrate levels were evident in the Zn-stressed HepG2 mitochondria as opposed to control mitochondria (Fig. 2,I). In addition, higher levels of α-KG were also present in the HepG2 mitochondria challenged with Zn. These chromatographic studies confirmed the inability of theZn-stressed HepG2 mitochondria to appropriately meta-bolize citrate. Higher amounts of citrate and α-KG in the Zn-stressed mitochondria indicated that this divalentmetal may affect various enzymes contributing to theTCA cycle. Aconitase, an enzyme essential in the isomerization of citrate was evaluated (Fig. 2, II). Amarked decrease in the expression of this enzyme wasobserved in the mitochondria isolated from the Zn- stressed cells. However, similar levels were evident in thecytoplasmic fraction. The purity of these cellular fractionswas ascertained with F-actin and VDAC immunoblots(Fig. 2, III and IV). As Zn was triggering oxidativestress, the cells were also exposed to H2O2 in order toassess the similarities/differences between these two insults. KGDH is a key enzyme in the TCA cycleinvolved in the production of NADH, a consequence ofthe oxidative decarboxylation of α-KG. KGDH activitywas ascertained by native PAGE using NAD+, CoA and
α-KG. The KGDH in the Zn-stressed mitochondria displayed lower activity in comparison with the control(Fig. 3, I). A decreased KGDH activity was alsoobserved in cells exposed to H2O2. To evaluate the amount of protein associated with the activity stain, the bands wereexcised and treated for 2D immunoblot analysis. The intensity of the bands did not appear to vary significantly among the three conditions (Fig. 3, II). It is importantto note that H2O2 and Zn appeared to elicit a similarresponse in regard to KGDH. As Zn seemed to interferewith the KGDH-mediated production of NADH, other enzymes critical in the aerobic production of NADHwere also examined. NAD+-ICDH activity was severelyhampered in HepG2 cells exposed to Zn (Fig. 3, III). In contrast, the activity of this enzyme seemed to be unaffected in the control cells. To assess the influence ofZn on oxidative metabolism further, two key enzymes involved in oxidative phosphorylation, SDH and Cyt COx, were probed. In-gel activity staining of SDH pointed to the diminished capacity of this enzyme to metabolizesuccinate in the Zn-stressed HepG2 cells (Fig. 3, IV).Similarly, the amount of diaminobenzidene precipitated at the site of Cyt C Ox activity was severely impeded in theZn-stressed HepG2 cells (Fig. 3, V). The inhibition ofSDH and Cyt C Ox was also confirmed in cells exposedto H2O2. Since NAD+-ICDH was diminished in theZn-challenged cells that were characterized by higheramounts of α-KG; it was decided to evaluate the statusof NADP+-ICDH, an enzyme that produces α-KG. This enzyme has also been localized in the mitochondriawhere it has been shown to play a pivotal role in main-taining a reductive environment. Enhanced activity of thecytoplasmic NADP+-ICDH was observed in the cellsexposed Zn (Fig. 4, I). Mitochondrial NADP+-ICDH exposed to Zn and H2O2 also showed an increase in activity (Fig. 4, II). The activity bands were excised andelectrophoresed by the 2D-PAGE technique. Subsequent silver staining and immunoblot analysis revealed moreprotein associated with the activity band from the cells
Table 1.Oxidized lipids and proteins in control and Zn-exposed HepG2 cells
Treatment Oxidized lipids (μmola/4 × 106cells)Oxidized protein (pmolb/4 × 106 cells)
Control0.050 ± 0.0160.339 ± 0.096c Zn stressed0.156 ± 0.0931.209 ± 0.058cn = 3, mean ± SD.a Malondialdehyde equivalents (μmol).b Carbonyl equivalents (ρmol).c Denotes a statistically significant difference in comparison with control (P ≤ 0.05).
rial activity: HepG2 cells were grown on a glass cover-
slip until minimal confluency and then exposed to (A)
citrate and (B) Zn-citrate for 24 h. The mitochondria
were visualized with anti-VDAC and Rhodamine B.
Scale bar: 10 μm. This figure is available in colour
online at www.interscience.wiley.com/journal/jat
Immunofluorescent detection of mitochond-
ZINC TOXICITY ON ATP PRODUCTION179Copyright © 2007 John Wiley & Sons, Ltd.J. Appl. Toxicol. 2008; 28: 175–182DOI: 10.1002/jatthat were treated with Zn and H2O2, respectively (Fig. 4,III and IV). As the activity and expression of severalTCA cycle and ETC enzymes were perturbed by the presence of Zn, the ability of the mitochondria to produceATP was assessed by HPLC. Incubation of control andZn-stressed mitochondria in the presence of 1mM citrateand 0.5mM ADP revealed diminished ATP production inthe Zn-stressed HepG2 cells. While control cells had a prominent ATP signal at 6.5min, the signal at 6min attributable to ADP, was more pronounced in the Zn-stressed cells (Fig. 5). Thus, indicating the diminished ability of the Zn-challenged cells to consume ADP andproduce ATP. The data in this report point to the negative impact ofZn on mitochondrial ATP production. Zn imposed an oxidative stress, severely impeding various TCA cycle enzymes and limiting energy production via oxidativephosphorylation. Toxic levels of Zn have been implicated in promoting oxidative stress, however; the molecular mechanism underlying this process is still unclear. In thisstudy a marked reduction of the mitochondrial membranepotential in HepG2 cells challenged with Zn was observed. It is therefore not unlikely that this transitionelement may interfere with the mitochondrial enzymes involved in the generation of the proton gradient. Indeed,this soft acid metal can form strong interactions withenzymatic imidazole and sulphydryl groups. Numerousenzymes involved in the aerobic production of ATPcontain active sites rich in these side chains (Quig, 1998).Although these interactions form the basis of the structural and functional role of zinc in biology, higherlevels may impede the action of enzymes which rely on
then analysed. (A) Citrate cultures. (B) Zn-citrate cultures. (II) Immunoblot analysis of ACN in HepG2 cells exposed
to citrate and Zn-citrate cultures. (A) Citrate stressed cells. (B) Zn-citrate stressed cells. Std corresponds to an ACN
(porcine heart) purchased from Sigma. (III, IV) The purity of cellular fractions and equal loading were ascertained
by immunoblot assays. (A) Cytoplasm from citrate culture. (B) Cytoplasm from Zn-citrate culture. (C) Mitochondrial
fraction from citrate culture. (D) Mitochondrial fraction from Zn-citrate culture. (III) Immunoblot for VDAC. (IV)
Immunoblot for F-Actin
(I) HPLC analysis of citrate metabolism in HepG2 cells. Mitochondria were incubated for 1 h at 37 °C and
180J. LEMIRE ET AL.Copyright © 2007 John Wiley & Sons, Ltd.J. Appl. Toxicol. 2008; 28: 175–182DOI: 10.1002/jatcatalytic thiols. The mitochondrial ACN seemed to be a potent target of Zn toxicity. The disruption of the expres-sion of this enzyme may be due to the ability of Zn toperturb the Fe-S cluster necessary for catalysis. The proper functioning of mitochondrial ACN is critical to thecommitment of citrate to anaplerosis and ATP produc- tion. The chromatographic data revealed the inability ofZn-stressed HepG2 mitochondria to metabolize this tricarboxylic acid effectively. A much higher signal con-
Mitochondrial fraction isolated from citrate exposed
cells. (B) Mitochondrial fraction isolated from Zn-citrate
exposed cells. (C) Mitochondrial fraction isolated from
H2O2 exposed cells. (I) In gel detection of α-KGDH ac-
tivity. Bands were quantified using Scion Imaging for
Windows. (II) 2D SDS-PAGE of α-KGDH; BN-PAGE activ-
ity bands followed by immunoblot analysis. Std corre-
sponds to an α-KGDH (Porcine heart) purchased from
Sigma. (III) In gel detection of NAD+-ICDH in the cyto-
plasm. (IV, V) In gel activity staining for SDH and Cyt C
Ox in mitochondria isolated from (A) citrate culture,
(B) H2O2 and citrate culture, (C) Zn-citrate culture, (IV)
SDH, (V) Cyt C Ox
Activity and expression of α-KGDH. (A)
ICDH in HepG2 cells exposed to (A) citrate, (B) Zn-
citrate stress, (C) H2O2 and citrate stress. (I) In gel
detection of NADP+-ICDH in the cytoplasm. (II) In gel
detection of NADP+-ICDH in the mitochondria. (III) 2D
SDS PAGE on NADP+-ICDH in the mitochondria. (IV) 2D
immunoblot of NADP+-ICDH in the mitochondria
Activity stain and immunoblot analysis of
firmed the diminished capacity of ACN to isomerize citrate to isocitrate. Interestingly Zn did not hamper theexpression of the cytoplasmic ACN enzyme. This observ-ation is consistent with the role of the cytoplasmic ACN in initiating transcriptional events (Crichton et al., 2002;
ATP and ADP levels in citrate and Zn-citrate stressed
HepG2 cells. Mitochondria were incubated for 1 h at
37 °C. (A) Citrate culture. (B) Zn-citrate culture. ATP
was monitored at 254 nm
HPLC chromatographs displaying relative
ZINC TOXICITY ON ATP PRODUCTION181Copyright © 2007 John Wiley & Sons, Ltd.J. Appl. Toxicol. 2008; 28: 175–182 DOI: 10.1002/jatMailloux et al., 2006a). Zn-mediated disruption of Fe-S dependent enzymes was not limited to ACN. Fe-Scluster-dependent enzymes such as SDH and Cyt C Oxalso succumbed to the toxicity of Zn. The diminished activity of these enzymes would negate the production ofmitochondrial ATP. The inability to produce ATP via O2-dependent metabolism would have dire consequences, asthis nucleotide is required to drive cellular processes.Indeed, HPLC analysis pointed to a decrease in ATPproduction in Zn-stressed mitochondria. These data areconsistent with the lowered mitochondrial membranepotential as revealed by fluorescence microscopic assays(Jiang et al., 2001). In addition the H2O2-stressed HepG2 cells were also characterized with the lowered activityof the two oxidative phosphorylation enzymes. Metal-mediated disruption of Fe-S cluster dependent TCAcycleenzymes has been shown previously (Mailloux et al., 2006a; Middaugh et al., 2005). In fact, it has recentlybeen shown that the bioavailability of Zn is tightlyregulated within the intracellular environment. Small increments in Zn levels leads to the activation of a hyper-sensitive transcriptional switch ZntR, which modulatesthe availability of Zn in the femtomolar range (Changela et al., 2003; Outten and O’Halloran, 2001). The tight regulation of intracellular Zn levels may be due to itsaffinity for biological active sites. Indeed this divalent cation has been implicated in oxidative stress and numer-ous pathologies (Barceloux, 1999; Fosmire, 1990; Koh et al., 1996). Competition for Fe-S binding sites may displace free Fe and contribute to oxidative tension within the cell. This divalent cation has also been reported to interact with thiol groups of antioxidants suchas glutathione (Nappi and Vass, 1997; Sipos et al., 2002).This would inevitably lead to the depletion of reduced glutathione in the intracellular environment thus contrib-uting to oxidative tension. In addition Zn-mediated inhi- bition of SDH and Cyt C Ox would certainly contributeto e− leakage and ROS production in the mitochondria.Mitochondrial activity generates a highly oxidative envi- ronment as it is the site of oxidative phosphorylation(Martin et al., 1996; Shigenaga et al., 1994). In this studytwo critical NAD+-dependent decarboxylating enzymes of the TCA cycle, ICDH and KGDH, exhibited lower activ-ity and expression within the Zn-stressed mitochondria.Interestingly, no reduction in expression of KGDH wasobserved, even though the activity was considerably less.This observation may be attributable to the redox sensi-tive lipoic acid residue on the E2-subunit of this enzyme.Zn may interact with the available thiols on the enzymerendering it inactive. In addition, an oxidative environ-ment may promote the oxidation of the thiols. The inhibi-tion of these two decarboxylation enzymes was alsoevident in the presence of H2O2. Hence, it is tempting to postulate that the reduction in activity of this enzymewithout a concomitant diminution in expression may con-tribute to the reduction of the oxidative tension. ROSproduction during metal toxicity is a well documentedphenomenon. NADPH-dependent enzymes such as cyto-plasmic ICDH, malic enzyme and glucose-6-phosphatedehydrogenase have recently been identified as key con-tributors to the antioxidant system (Beriault et al., 2007). In the Zn-stressed HepG2 cells, the cytoplasmic NADP+-ICDH had much higher activity. Mitochondrial NADP+-ICDH also displayed higher activity in the Zn treatedcells. These observations point to an attempt by the cellto cope with oxidative stress by producing large amountsof NADPH. NADPH is crucial in providing the cellswiththe reducing power necessary to regenerate antioxidants such as catalase, SOD and glutathione (Arivazhaganet al., 2000; Fang et al., 2002). Thus, the enhanced pro-duction of NADPH would ensure the maintenance of thereducing environment necessary for scavenging ROS. This observation may also account for the accumulation of α-KG. α-KG is known as a powerful antioxidant and utilized in the prevention of inflammation during surgery(Kjellman et al., 1995; Velvizhi et al., 2002). Thismetabolite has been shown to accumulate as a conse-quence of metal toxicity (Zatta et al., 2000). The accu-mulation of α-KG observed in this instance may be dueto the inhibition of KGDH by Zn and an overactive mitochondrial NADP+-ICDH. Hence, this α-keto acid may serve as an important tool against oxidative stress.The modulation of these TCA cycle enzymes would serve as a natural antioxidant defense system whichwould aid in attenuating oxidative stress. Therefore, it isnot unlikely that the TCA cycle can serve to regulate the oxidative milieu within the mitochondria by modulating key enzymes. The oxidation and reduction of the lipoicacid residue on the E2-subunit may serve such a function.In conclusion, this study demonstrates that Zn exertsits toxic influence by creating an oxidative environment and interfering with several critical enzymes that partici-pate in the TCA cycle and ETC. This inhibition severelyrestricts the ability of the cell to generate ATP, an ingre-dient essential for the proliferation and survival of thecell. This is the first demonstration of the upregulation ofthe mitochondrial NADP+-ICDH as a consequence ofoxidative stress induced by Zn toxicity. Thus it is quite likely that an ineffective TCA cycle and oxidativephosphorylation may be important contributing factors invarious Zn-triggered pathologies.Acknowledgements—This work was supported by funding from Industry Canada.
Abordo EA, Minhas HS, Thornalley PJ. 1999. Accumulation of alpha-oxoaldehydes during oxidative stress: a role in cytotoxicity. Biochem.Pharmacol. 58: 641–648. Arivazhagan P, Thilakavathy T, Panneerselvam C. 2000. Antioxidantlipoate and tissue antioxidants in aged rats. J. Nutr. Biochem. 11:122–127.
182J. LEMIRE ET AL.Copyright © 2007 John Wiley & Sons, Ltd.J. Appl. Toxicol. 2008; 28: 175–182DOI: 10.1002/jat Aydin S, Yargicoglu P, Derin N, Aliciguzel Y, Abidin I, Agar A. 2005. The effect of chronic restraint stress and sulfite on visual evokedpotentials (VEPs): relation to lipid peroxidation. Food Chem. Toxicol.43: 1093–1101.Barceloux DG. 1999. Zinc. J. Toxicol. Clin. Toxicol. 37: 279–292.Beriault R, Hamel R, Chenier D, Mailloux RJ, Joly H, Appanna VD. 2007. The overexpression of NADPH-producing enzymes countersthe oxidative stress evoked by gallium, an iron mimetic. Biometals20: 165–176. Bradford MM. 1976. A rapid and sensitive method for the quantitationof microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248–254. Canzoniero LM, Turetsky DM, Choi DW. 1999. Measurement of intracellular free zinc concentrations accompanying zinc-inducedneuronal death. J. Neurosci. 19: RC31.Changela A, Chen K, Xue Y, Holschen J, Outten CE, O’Halloran TV,Mondragon A. 2003. Molecular basis of metal-ion selectivity andzeptomolar sensitivity by CueR. Science 301: 1383–1387. Coleman JE. 1992. Zinc proteins: enzymes, storage proteins, transcrip-tion factors, and replication proteins. Annu. Rev. Biochem. 61: 897–946. Costello LC, Liu Y, Franklin RB, Kennedy MC. 1997. Zinc inhi-bition of mitochondrial aconitase and its importance in citratemetabolism of prostate epithelial cells. J. Biol. Chem. 272: 28875–28881. Crichton RR, Wilmet S, Legssyer R, Ward RJ. 2002. Molecular andcellular mechanisms of iron homeostasis and toxicity in mammaliancells. J. Inorg. Biochem. 91: 9–18.Dinely KE, Votyakova TV, Reynolds IJ. 2003. Zinc inhibition ofcellular energy production: implication for mitochondria andneurodegeneration. J. Neurochem. 85: 563–570.Dumont A, Hehner SP, Hofmann TG, Ueffing M, Droge W, Schmitz ML. 1999. Hydrogen peroxide-induced apoptosis is CD95-independent, requires the release of mitochondria-derived reactiveoxygen species and the activation of NF-kappaB. Oncogene 18: 747–757. Fang YZ, Yang S, Wu G. 2002. Free radicals, antioxidants, and nutrition. Nutrition 18: 872–879.Fernie AR, Carrari F, Sweetlove LJ. 2004. Respiratory metabolism:glycolysis, the TCA cycle and mitochondrial electron transport. Curr.Opin. Plant Biol. 7: 254–261.Fosmire GJ. 1990. Zinc toxicity. Am. J. Clin. Nutr. 51: 225–227. Gazaryan IG, Krasnikov BF, Ashby GA, Thorneley RN, Kristal BS,Brown AM. 2002. Zinc is a potent inhibitor of thiol oxidoreductase activity and stimulates reactive oxygen species production bylipoamide dehydrogenase. J. Biol. Chem. 277: 10064–10072.Hambidge M. 2000. Human zinc deficiency. J. Nutr. 130: 1344S–1349S. Jiang D, Sullivan PG, Sensi SL, Steward O, Weiss JH. 2001. Zn(2+)induces permeability transition pore opening and release of pro-apoptotic peptides from neuronal mitochondria. J. Biol. Chem. 276:47524–47529.Kjellman U, Bjork K, Ekroth R, Karlsson H, Jagenburg R, Nilsson F,Svensson G, Wernerman J. 1995. Alpha-ketoglutarate for myocardial protection in heart surgery. Lancet 345: 552–553. Koh JY, Suh SW, Gwag BJ, He YY, Hsu CY, Choi DW. 1996. Therole of zinc in selective neuronal death after transient global cerebralischemia. Science 272: 1013–1016. Laemmli UK. 1970. Cleavage of structural proteins during the assem-bly of the head of bacteriophage T4. Nature 227: 680–685.Mailloux RJ, Appanna VD. 2007. Aluminum toxicity triggers thenuclear translocation of HIF-1alpha and promotes anaerobiosis inhepatocytes. Toxicol. In Vitro 21: 16–24. Mailloux RJ, Hamel R, Appanna VD. 2006a. Aluminum toxicity elicits a dysfunctional TCA cycle and succinate accumulation inhepatocytes. J. Biochem. Mol. Toxicol. 20: 198–208. Mailloux RJ, Singh R, Appanna VD. 2006b. In-gel activity staining ofoxidized nicotinamide adenine dinucleotide kinase by blue nativepolyacrylamide gel electrophoresis. Anal. Biochem. 359: 210–215.Manev H, Kharlamov E, Uz T, Mason RP, Cagnoli CM. 1997. Char- acterization of zinc-induced neuronal death in primary cultures of ratcerebellar granule cells. Exp. Neurol. 146: 171–178. Martin GM, Austad SN, Johnson TE. 1996. Genetic analysis of ageing:role of oxidative damage and environmental stresses. Nat. Genet. 13:25–34.Middaugh J, Hamel R, Jean-Baptiste G, Beriault R, Chenier D,Appanna VD. 2005. Aluminum triggers decreased aconitase activityvia Fe-S cluster disruption and the overexpression of isocitrate dehydrogenase and isocitrate lyase: a metabolic network mediatingcellular survival. J. Biol. Chem. 280: 3159–3165.Nappi AJ, Vass E. 1997. Comparative studies of enhanced iron- mediated production of hydroxyl radical by glutathione, cysteine,ascorbic acid, and selected catechols. Biochim. Biophys. Acta 1336:295–302.Outten CE, O’Halloran TV. 2001. Femtomolar sensitivity ofmetalloregulatory proteins controlling zinc homeostasis. Science 292: 2488–2492. Piao F, Yokoyama K, Ma N, Yamauchi T. 2003. Subacute toxic effectsof zinc on various tissues and organs of rats. Toxicol. Lett. 145: 28–35. Prasad AS. 1996. Zinc deficiency in women, infants and children. J. Am. Coll. Nutr. 15: 113–120. Quig D. 1998. Cysteine metabolism and metal toxicity. Altern. Med.Rev. 3: 262–270. Religa D, Strozyk D, Cherny RA, Volitakis I, Haroutunian V, WinbladB, Naslund J, Bush AI. 2006. Elevated cortical zinc in Alzheimer disease. Neurology 67: 69–75. Samizo K, Ishikawa R, Nakamura A, Kohama K. 2001. A highly sensitive method for measurement of myosin ATPase activity byreversed-phase high-performance liquid chromatography. Anal.Biochem. 293: 212–215. Sandstead HH. 1991. Zinc deficiency. A public health problem? Am. J. Dis. Child. 145: 853–859.Schagger H, von Jagow G. 1991. Blue native electrophoresis for isolation of membrane protein complexes in enzymatically activeform. Anal. Biochem. 199: 223–231.Sensi SL, Yin HZ, Weiss JH. 1999. Glutamate triggers preferentialZn2+ flux through Ca2+ permeable AMPA channels and consequentROS production. Neuroreport 10: 1723–1727. Shannon JE. 1978. Tissue culture viability assays — a review of theliterature. Cryobiology 15: 239–241.Shigenaga MK, Hagen TM, Ames BN. 1994. Oxidative damage andmitochondrial decay in aging. Proc. Natl Acad. Sci. USA 91: 10771– 10778. Sipos K, Lange H, Fekete Z, Ullmann P, Lill R, Kispal G. 2002.Maturation of cytosolic iron-sulfur proteins requires glutathione.J. Biol. Chem. 277: 26944–26949. Steinebach OM, Wolterbeek HT. 1993. Effects of zinc on rat hepatomaHTC cells and primary cultured rat hepatocytes. Toxicol. Appl.Pharmacol. 118: 245–254. Takeda A, Minami A, Takefuta S, Tochigi M, Oku N. 2001. Zinc homeostasis in the brain of adult rats fed zinc-deficient diet. J. Neurosci. Res. 63: 447–452. Velvizhi S, Dakshayani KB, Subramanian P. 2002. Protective influences of alpha-ketoglutarate on lipid peroxidation and antioxidant status inammonium acetate treated rats. Indian J. Exp. Biol. 40: 1183–1186.Walsh CT, Sandstead HH, Prasad AS, Newberne PM, Fraker PJ. 1994. Zinc: health effects and research priorities for the 1990s. Environ. Health Perspect. 102 (Suppl 2): 5–46. Zatta P, Lain E, Cagnolini C. 2000. Effects of aluminum on activity ofKrebs cycle enzymes and glutamate dehydrogenase in rat brainhomogenate. Eur. J. Biochem. 267: 3049–3055. Download full-text