Zinc causes loss of membrane potential and elevates reactive oxygen species in rat brain mitochondria

Mitochondrion 5 (2005) 55–65 www.elsevier.com/locate/mito

Zinc causes loss of membrane potential and elevates reactive oxygen species in rat brain mitochondria Kirk E. Dineley1, Lauren L. Richards, Tatyana V. Votyakova, Ian J. Reynolds* Department of Pharmacology, University of Pittsburgh, W1351 Biomedical Science Tower, Pittsburgh, PA 15261, USA Received 26 July 2004; received in revised form 11 November 2004; accepted 18 November 2004

Abstract Emerging evidence suggests that Zn2C may impair neuronal metabolism. We examined how Zn2C affects the activity of isolated brain mitochondria fueled with glutamateCmalate, succinate or glycerol 3-phosphate. Submicromolar levels of Zn2C dissipated membrane potential and inhibited oxygen utilization in all three substrate conditions. Zn2C-induced depolarization was reversed by the membrane-impermeant metal chelator, EGTA, and was inhibited by uniporter blockade. Cyclosporin A did not block Zn2C-induced depolarization. Added Zn2C increased accumulation of reactive oxygen species (ROS) in glutamateC malate or glycerol 3-phosphate conditions, but inhibited succinate-supported ROS accumulation. These results show that Zn2C blocks mitochondrial function in all physiologically relevant substrate conditions. q 2004 Elsevier B.V. and Mitochondria Research Society. All rights reserved. Keywords: ROS; Spectrofluorometry; Permeability transition; Calcium uniporter; Excitotoxicity; Neurodegeneration

1. Introduction Mitochondria are critical for maintaining neuronal homeostasis. In addition to producing large quantities of ATP required for neuronal function, mitochondria are important regulators of intracellular calcium, the Abbreviations used: CsA, cyclosporin A; DJm, mitochondrial membrane potential; ETC, electron transport chain; GCM, glutamate and malate; G3P, glycerol 3-phosphate; MPT, mitochondrial permeability transition; ROS, reactive oxygen species; [Zn2C], free Zn2C concentration. * Corresponding author. Tel.: C1 412 648 2134, fax: C1 412 624 0794 E-mail address: [email protected] (I.J. Reynolds). 1 Department of Biology, Francis Marion University, Florence, SC 29501, USA.

major site of reactive oxygen species (ROS) production, and are thought to be key activators of programmed cell death. While mitochondrial uptake of calcium as well as limited production of ROS appear necessary for normal cellular functions, either in excess contributes to neuronal injury. Under such stresses, mitochondria can initiate cell death by releasing pro-apoptotic factors such as cytochrome C and apoptosis inducing factor (AIF). Given their central role in both normal and pathological cellular processes, it is not surprising that mitochondrial disruption is considered a critical event in various neurodegenerative schemes, including models of stroke, amyotrophic lateral sclerosis, and Parkinson’s (for reviews, see Nicholls and Budd, 2000; Zamzami and Kroemer, 2001).

1567-7249/$ - see front matter q 2004 Elsevier B.V. and Mitochondria Research Society. All rights reserved. doi:10.1016/j.mito.2004.11.001

56

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65

Zinc is present in cells at concentrations up to approximately 200 mM. The vast majority of cellular zinc is bound to intracellular sites such as proteins, consequently there is very little free ionic zinc (Zn2C) in the cytoplasm. Tight regulation of intracellular free Zn2C is necessary, because excessive elevation of free zinc is toxic (Weiss et al., 2000). Live cell estimation of free zinc is difficult (Dineley et al., 2002), however it has been reported that free zinc in the range of several hundred nanomolar kills neurons (Canzoniero et al., 1999). While Zn2C can enter neurons through membrane channels and transporters, increasing evidence suggests that zinc-mediated neuronal death is caused by liberating zinc from intracellular sites. For instance, several reports suggest that oxidantmediated neuronal injury involves mobilization of excessive amounts of zinc (Cuajungco and Lees, 1998; Aizenman et al., 2000), presumably from metalloproteins (Maret and Vallee, 1998). Other reports suggest that zinc causes ROS accumulation in live cells, possibly from multiple mechanisms including inhibition of mitochondria (Sensi et al., 1999) and activation of NADPH oxidase (Noh et al., 1999). Precisely how excess [Zn2C] i kills neurons remains unclear, but increasing evidence indicates that Zn2C impedes cellular energy production (reviewed by Dineley et al., 2003). The impact of Zn2C on the activity of isolated mitochondria has been explored with varying results. Zn2C probably inhibits electron transport at the bc1 complex (Lorusso et al., 1991; Link and von Jagow, 1995; Berry et al., 2000), but there may be additional sites of inhibition in the electron transport chain (Skulachev et al., 1967; Nicholls and Malviya, 1968) or the tricarboxylic acid cycle (Brown et al., 2000). Also, there is conflicting evidence with respect to Zn2C-induced permeability transition (Wudarczyk et al., 1999; Brown et al., 2000; Jiang et al., 2001). A number of factors could contribute to these disparate conclusions including assay conditions using ion versus sucrose-based recording solutions, or substantial variations in the amount of Zn2C used (from w10K9 to O10K5 M). In any case, because the majority these studies used nonneural mitochondria they may be of limited relevance to neurodegeneration. The present study employs fluorometric and polarographic techniques to directly determine how

Zn2C alters membrane potential (Djm), O2 utilization, and ROS accumulation in isolated brain mitochondria fueled by three different substrate conditions. Because these substrates enter the electron transport chain at different sites, this approach may provide insight regarding the important site(s) of Zn2C inhibition. More importantly, this approach reveals what alternate catabolic pathways, if any, mitochondria may use when challenged by Zn2C. We found that [Zn2C] in the approximate range of 100– 200 nM inhibits O2 utilization, dissipates Djm, and alters ROS accumulation in a substrate dependent fashion. Together, these results show that neurotoxic concentrations of zinc profoundly inhibit mitochondrial activity and suggest several important mechanisms for Zn2C-mediated neuronal death. Some of these results have been presented previously in an abstract (Dineley et al., 2001).

2. Materials and methods 2.1. Isolation of rat brain mitochondria All procedures using rats were approved by the Institutional Animal Care and Use Committee of the University of Pittsburgh and are consistent with guidelines provided by the National Institutes of Health. Rat brain synaptosomal mitochondria were isolated from the cortex of adult Sprague Dawley rats by conventional differential centrifugation as described by Rosenthal et al. (1987) with minor modifications. Following removal, the brain was homogenized at 0–2 8C in an isolation buffer (IB) containing (in mM): 225 mannitol; 75 sucrose; 5 HEPES; 0.5 EDTA; with 1 mg/mL BSA and adjusted to pH 7.35 with KOH. The supernatant was collected after two rounds of centrifugation at 2000g (three minutes each), then was then centrifuged for 10 min at 10,500g to yield the first mitochondrial pellet. The first pellet was resuspended in IB containing digitonin (0.013%) and centrifuged again. The supernatant was discarded, the pellet was resuspended in IB with 10 mM EGTA and no EDTA, and the final pellet was retrieved after a final round of centrifugation. Supernatant was discarded, and the pellet collected in w100 mL of the EGTA buffer, typically yielding a final concentration of 20–30 mg protein/mL as

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65

determined by the Bradford method (Bradford, 1976). Mitochondria prepared this way typically showed a respiratory control index of 6–8 when fueled by glutamate with malate, and were active for 5–6 h as determined by their ability to maintain a transmembrane potential in the presence of oxidizable substrates. 2.2. Measurements of Djm, H2O2 accumulation, and O2 utilization in isolated mitochondria Fluorescence measurements were performed in a Shimadzu RF5301 spectrofluorometer in a stirred cuvette maintained at 37 8C. Mitochondria (0.2 mg/mL) were added to 1.5 mL volume of standard incubation buffer, which contained (in mM): 125 KCl; 2 K2HPO4; 5 MgCl2; 5 HEPES and 0.01 EGTA. Three different substrate conditions were used: (i) 5 mM glutamate with 5 mM malate, (ii) 5 mM succinate, or (iii) 10 mM glycerol 3-phosphate. pH was adjusted to 7.0 with KOH. Hydrogen peroxide production was measured using Amplex Red (4 mM). Amplex Red is a horseradish peroxidase substrate that generates a fluorescent product (resorufin) through H2O2-mediated oxidation. Thus, increased fluorescence indicates increased levels of H2O2. Measurements were carried out at excitation/emission wavelengths of 560 and 590 nm, respectively. Mitochondrial transmembrane potential (Djm) was estimated using the cationic, lipophilic fluorophore safranine O (2.5 mM) (Akerman and Wikstrom, 1976). The high transmembrane potential of energized mitochondria causes the accumulation and aggregation of safranine O. Potential-induced dye aggregation decreases safranine O fluorescence. Conversely, dissipation of Djm (e.g. with FCCP) results in dye release, de-aggregation, and increased fluorescence. Thus, the magnitude of safranine O fluorescence is inversely related to Djm. Safranine O was excited at 495 nm and emitted light was collected at 590 nm. None of the added reagents affected signal from either Amplex Red or safranine O. O2 measurements were performed with a standard Clark-type electrode apparatus. The incubation medium used in polarography experiments was supplemented with ADP (500 mM). Succinate polarography experiments included rotenone (500 nM) to preclude contribution of complex I activity.

57

2.3. Data analysis Mitochondrial depolarization was analyzed in one of two ways. For analysis of single-bolus Zn2C concentration-response data (as presented in Fig. 1), mitochondria were completely depolarized with FCCP (150 nM) at the end of the experiment. This change in fluorescence was used as a standard of maximum depolarization, against which all other previous points in the recording were expressed as a percentage. For Djm experiments testing inhibitors such as in Fig. 2, we calculated the area under the curve (AUC) between the first addition of Zn2C and immediately before the addition of EGTA for each trace. Because sequential additions of small amounts of Zn2C generated variability in the onset, slope and extent of depolarization, calculations of AUC presented a convenient summary approach. Accordingly, Zn2C alone or Zn2C with inhibitor was normalized to AUC of traces from untreated mitochondria from the same preparation. In Amplex Red and polarography experiments, slopes for each trace were calculated. Values obtained from experimental conditions were normalized to slopes generated by untreated mitochondria from the same preparation. All experiments were repeated at least three times on mitochondria from at least three different preparations, and p%0.05 was considered significant. GraphPad Prism 3.02 (San Diego CA) was used for all statistical analyses. 2.4. Controlling [Zn2C]. We used the web-based program MAXCHELATOR (http://www.stanford.edu/~cpatton/maxc.html) to estimate free Zn2C. EGTA (10 mM) was added fresh to incubation medium prior to experiments. At pH 7.0, 37 8C and ionic strength 0.140 M, the Kd for EGTA and Zn2C is 4.49 nM. EGTA was present at 10 mM, and Zn2C was usually added at 5, 10 and 20 mM amounts. According to MAXCHELATOR, this would yield free Zn2C concentrations of approximately 5 nM, 200 nM or 10 mM, respectively. Thus we generally assayed mitochondrial activity in three basic conditions: very little free zinc, pathophysiologically relevant free zinc, and supraphysiological free zinc. It is important to note that the above calculations likely overestimate free zinc, because the calculation considers only Zn2C and EGTA; it does not include

58

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65

Fig. 1. Zn2C depolarizes isolated brain mitochondria. Mitochondria were added to incubation medium containing the potentiometric dye safranine O (2 mM) and either 5 mM glutamateC5 mM malate (A), 5 mM succinate (C) or 10 mM glycerol 3-phosphate (E), resulting in the formation of a transmembrane gradient (y axis for A, C, and E is raw Safranine O fluorescence units). At about 2.5 min, Zn2C (5, 10 or 20 mM) was added (solid traces). EGTA (1 mM) was added at about 10 min. Complete depolarization was caused by FCCP (150 nM) at about 12 min. Dashed recordings are from untreated mitochondria, and show extent of basal depolarization over the experimental time course. In B, D, and F summary data are provided for each substrate condition. For experiments when Zn2C was added, first bar shows depolarization induced by zinc, and the second bar (CEGTA) shows EGTA-induced recovery. In all substrate conditions, EGTA reversed depolarization caused by 10 mM added Zn2C. Depolarization is expressed as percentage of FCCP-induced depolarization observed in the same trace. * columns differ significantly according to students two-tailed t-test, p!0.05. Bars represent meanGSE; for each condition experiments were repeated at least three times on mitochondria from three different preparations.

buffering from other components in the medium such as phosphate, dicarboxylic acid substrates, and residual EDTA from the isolation procedure. 2.5. Reagents Amplex Red was obtained from Molecular Probes (Eugene OR). All other reagents were purchased from Sigma (St. Louis, MO).

3. Results In Fig. 1, mitochondria fueled by GCM (A), succinate (C), or G3P (E) quickly established and maintained Djm (dashed traces). Djm was abolished by adding the protonophore FCCP (150 nM) at

approximately 12 min. Added Zn2C (in boluses of 5, 10, or 20 mM) caused loss of Djm in a concentration-dependent manner (solid traces). A membrane impermeant Zn 2C chelator (EGTA, 1 mM) always resulted in at least a partial restoration of Zn2C-induced loss of Djm, although mitochondria using succinate or G3P showed less resilience to higher added [Zn2C] compared to those using GCM. Interestingly, the membrane permeant Zn2C chelator TPEN was no more effective than EGTA in reversing Zn2C-induced loss of Djm (data not shown), suggesting that dissipation of Djm did not require entry of Zn2C into mitochondria. It has been suggested that Zn2C gains access to the mitochondrial matrix (Brierley and Knight, 1967), possibly via the Ca2C uniporter (Saris and Niva, 1994). It has also been reported that Zn2C induces

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65

59

Fig. 2. Ruthenium red, but not cyclosporin A partially protects mitochondria from Zn2C-induced depolarization. A, C, and E. Mitochondria were added to incubation medium containing substrates, as in Fig. 1, and Zn2C was added stepwise in 2.5 mM aliquots as indicated. Solid traces represent experiments without inhibitor, while dotted and dashed lines correspond to recordings where ruthenium red (200 nM) or CsA (2 mM), respectively, was added at approximately 1.5 min. EGTA and FCCP were added as described in Fig. 1B, D and F. Summary data are provided for each substrate condition. Extent of depolarization is a function of the area under the safranine O trace (i.e. area under curve) between the first addition of Zn2C and then immediately before the addition of EGTA. *ZincCRR, but not ZincCCsA differed significantly (p!0.05) from Zinc, according to one way ANOVA followed by Bonferroni post test. Bars represent meanGSE; for each condition experiments were repeated at least three times on mitochondria from three different preparations.

mitochondrial permeability transition (MPT) (Wudarczyk et al., 1999; Jiang et al., 2001). We examined whether Zn 2C-induced loss of Dj m involved either of these two pathways. In experiments similar to those of Fig. 1, Zn2C was added stepwise (in 2.5 mM aliquots) to substrate-fueled mitochondria (Fig. 2). In all cases, Zn2C again caused loss of Djm that was both concentration-dependent and reversed by excess EGTA (1 mM). Ruthenium red (200 nM), an inhibitor of the Ca2C uniporter, partially blocked Zn2C-induced depolarization. In contrast, cyclosporin A (CsA), an inhibitor of mitochondrial permeability transition (MPT), did not protect mitochondria from Zn2C-induced depolarization (Fig. 2). These data suggest that Zn2C transport through the Ca2C uniporter may contribute to, but MPT is not critical for Zn2C-induced depolarization. Stimuli that cause depolarization can have opposite effects on mitochondrial oxygen consumption. Uncouplers such as FCCP and DNP collapse

mitochondrial potential by equilibrating the transmembrane proton gradient. In this scenario, electron transport and consequent oxygen consumption are accelerated and transmembrane proton pumping continues, but the transmembrane gradient cannot be sustained. Conversely, inhibitors of the electron transport chain such as rotenone and antimycin A block electron transport and O2 utilization, leading to gradual depolarization. Using polarography, we addressed both possibilities. Ten or 20 (but not 5) mM added Zn2C significantly inhibited oxygen consumption under all three substrate conditions (Fig. 3). In the cases of GCM and succinate, addition of EGTA (1 mM) markedly restored respiration. Interestingly, EGTA did not reverse the inhibitory effect of Zn2C in mitochondria using G3P. We are unable to explain this effect, except to speculate that mitochondrial GPDH requires some metal for activity that is removed by millimolar EGTA. The addition of high concentrations of EGTA did inhibit G3P supported

60

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65

Fig. 3. Zn2C inhibits O2 consumption in brain mitochondria. A, C, and E. Mitochondria were added to incubation medium containing substrates (as in Fig. 1), 500 mM ADP and, where indicated, 10 or 20 mM Zn2C. Dashed traces represent control conditions (no added Zn2C), and illustrate extent of O2 consumption over the experimental time course. Solid traces represent traces in which Zn2C was added to incubation medium (10 or 20 mM) before mitochondria. EGTA (1 mM) was added as indicated. B, D, and F. Summary data are provided for each substrate condition. Data are normalized to control experiments from same preparation of mitochondria. * indicated columns differed significantly according to students two-tailed t-test, p!0.05. Bars represent meanGSE; for each condition experiments were repeated at least three times on mitochondria from three different preparations.

respiration in the absence of Zn2C (data not shown), which would be consistent with this suggestion. The important conclusion from these data is that under all three substrate conditions, Zn2C inhibits O2 consumption. Considered together with Fig. 1, the data argue in favor of Zn2C inhibition of the electron transport chain as opposed to an uncoupling effect that might be observed if the inner membrane had been simply damaged or otherwise rendered more permeable. In support of this, Zn2C-inhibted mitochondria showed no marked increase in respiration when subsequently uncoupled with FCCP (150 nM) (Fig. 4). Reversing the experiment, i.e. uncoupling first with FCCP then adding Zn2C produced gave similar results (data not shown). Our data thus far suggest that Zn2C blocks electron transport, but the precise site(s) of Zn2C inhibition remains ambiguous (see Section 4). Inhibition of the electron transport chain can suppress or increase ROS accumulation, depending on substrate conditions and the site of inhibition

within the ETC (Turrens, 1997). For instance, brain mitochondria fueled by succinate generate high levels of superoxide through reverse electron flow from complex II (succinate dehydrogenase) to complex I (NADH dehydrogenase) (Cino and Del Maestro, 1989). As shown by Votyakova and Reynolds (2001) and Liu et al. (2002), ROS generation under these circumstances is blocked by complex I inhibition, relies on high transmembrane potential and is virtually extinguished by relatively slight depolarization. The opposite is true of mitochondria using glutamate and malate: low levels of ROS are produced under basal conditions, inhibition of complex I can increase ROS generation, and ROS production is relatively unaffected by slight changes in Djm. Because our results indicated that Zn2C inhibited the ETC, we investigated how Zn2C might impact mitochondrial ROS production. We used the Amplex Red system for detecting ROS in our system. The predominant ROS species produced by the electron transport chain is superoxide ð,OK 2 Þ.

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65

Fig. 4. Zn2C blocks O2 utilization in uncoupled mitochondria. Experimental conditions are similar to those used in Fig. 3. Dashed trace shows mitochondria respiring on GCM. At approximately 2 min, ADP supplies are exhausted, and FCCP (150 nM) is added at about 5 min. In the solid trace, 10 mM Zn2C was present at the start of the experiment, resulting in lower rate of respiration. At about 3.5 min, 10 mM Zn2C is added, resulting in almost complete inhibition. FCCP (150 nM) is added at w5 min, but respiration is not substantially accelerated.

However, it is believed that superoxide is quickly converted to H2O2 by manganese superoxide dismutase. In Fig. 5, we assayed mitochondrial H2O2 accumulation under all three substrate conditions. We

61

found that Zn2C affected ROS accumulation in a substrate-dependent manner. In mitochondria using GCM, added Zn2C (5 or 10 mM) augmented ROS production compared to controls by factors of 1.66G 0.09 and 1.97G0.22, respectively (meanGSE; p!0.05 in both cases). In mitochondria using G3P, 5 or 10 mM zinc increased the rate of ROS accumulation by factors of 1.55G0.07 and 1.49G0.16, respectively (meanGSE; p!0.05 in both cases). In mitochondria fueled by succinate, ROS accumulation normally occurs at a very rapid rate. 5 mM added Zn2C had little effect, whereas 10 and 20 mM significantly inhibited succinate-fueled ROS accumulation. With each substrate EGTA (1 mM) partially reversed the effects of Zn2C on ROS accumulation (data not shown). Our Djm data showed that ruthenium red but not cyclosporin A could partially block Zn2C-induced depolarization. We therefore tested whether these agents altered Zn 2C-induced changes in ROS accumulation under all three substrate conditions. Neither ruthenium red nor CsA altered Zn2C-induced changes in the rates of ROS accumulation for any of the three substrate conditions (Fig. 6). Our data therefore argue that Zn2C-mediated alterations in

Fig. 5. Zn2C alters ROS production in brain mitochondria. A, C and E. Mitochondria were added to incubation medium containing substrates (as in Fig. 1), the H2O2-sensitive dye Amplex Red (4 mM) and, where indicated, 5, 10 or 20 mM Zn2C. B, D and F present summary data for each substrate condition. Data are normalized to control experiments from same preparation of mitochondria. Bars represent meanGSE; for each condition experiments were repeated at least three times on mitochondria from three different preparations. * indicates values are significantly different from control values, p!0.05, by t-test.

62

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65

mitochondrial ROS accumulation do not involve uniporter uptake or MPT.

4. Discussion This report shows in isolated brain mitochondria that submicromolar [Zn2C] (i) rapidly dissipates Djm, (ii) inhibits oxygen consumption, and (iii) alters the rates of ROS accumulation in a substrate-dependent manner. Additionally, we show that an impermeant chelator can reverse Zn2C-induced loss of membrane potential and inhibitions of oxygen utilization. 4.1. What concentration of Zn2C is physiologically relevant?

Fig. 6. Zn2C-induced alterations in mitochondrial ROS production are unaffected by ruthenium red or CsA. Brain mitochondria were added to incubation medium containing substrates and Amplex Red (as in Fig. 4). Where indicated, Zn2C, ruthenium red (200 nM), and CsA (2 mM) were added to the medium before mitochondria. Rotenone (200 nM) was included in some GCM experiments for positive control. One way ANOVA analysis showed no difference between Zinc alone and ZincCruthenium red or ZincCCsA. Bars represent meanGSE; for each condition experiments were repeated at least three times on mitochondria from three different preparations.

Perhaps the most important consideration in this report pertains to the amount of Zn2C that is biologically relevant in neurodegeneration. Normally, [Zn2C]i is very low or even non-existent (Outten and O’Halloran, 2001). Injurious stimuli however may raise [Zn2C]i to several hundred nanomolar (Canzoniero et al., 1999). While our experimental paradigm used micromolar additions of Zn2C, it is important to note that recording solutions included 10 mM EGTA. If one assumes a dissociation constant of 4.27 nM for EGTA and Zn2C, the addition of 10 mM Zn2C for example would result in approximately 200 nM free Zn2C. This estimation is probably a generous one, as the calculation does not consider buffering from other species present in the recording solutions, including: phosphate (2 mM), dicarboxylic acid substrates (5 mM), and residual EDTA from the isolation procedure. The aggregate effect of these additional Zn2C chelators makes it impossible to precisely calculate [Zn2C]. If we consider only added Zn2C and EGTA, we can state with confidence however that 200 nM Zn2C represents an uppermost limit that is well within accepted estimations of pathophysiologically relevant [Zn2C]i. Careful consideration of [Zn2C] is especially important when relating our findings to the greater context of the existing literature, because many studies either used high [Zn2C] (O10 mM) or, when exploring the effects of lower (!10 mM) concentrations, employed no system for controlling [Zn2C] (Skulachev et al., 1967;

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65

Nicholls and Malviya, 1968; Wudarczyk et al., 1999; Jiang et al., 2001). 4.2. Where does Zn2C inhibit the ETC? The rapidity with which EGTA reversed Zn2C effects on Djm and O2, and that the membrane permeant chelator TPEN was no more effective than EGTA suggest that Zn2C inhibits primarily on the external face of the inner membrane. We are not certain of the precise site(s) of action within the electron transport chain however. The most parsimonious conclusion is that Zn2C inhibits a single site in the chain common to all three substrates pathways. Our results show that Zn2C caused loss of potential and decreased O2 consumption in brain mitochondria fueled by GCM, succinate, or G3P. All three substrate conditions use the ubiquinone/complex III segment, and Zn2C interference in this vicinity would be consistent with our data as well as with previous reports identifying the bc1 complex as a target for Zn2C (Lorusso et al., 1991; Link and von Jagow, 1995; Berry et al., 2000). Our results however do not exclude effects at additional sites. 4.3. Is zinc transported by the Ca2C uniporter? A related question pertains to possible Zn2C uptake by brain mitochondria, presumably through the Ca2C uniporter. As other neural Ca2C channels can also accommodate Zn2C (Sensi et al., 1997; Cheng and Reynolds, 1998), this is reasonable speculation, and indeed two studies are often cited as proof of uniporter mediated Zn2C influx (Brierley and Knight, 1967; Saris and Niva, 1994). We found that ruthenium red afforded partial protection from Zn2C-mediated depolarization. Our results are somewhat problematic in that further addition of Zn2C eventually overcame ruthenium red protection of Djm. Interestingly, ruthenium red is not thought to be a competitive antagonist of the uniporter (Reed and Bygrave, 1974), therefore its simple displacement by Zn2C seems unlikely. Zn2C uptake might be selflimiting however, as Zn2C rapidly abolishes the transmembrane gradient necessary to drive the uniporter in the first place. Published data in support of uniporter-mediated Zn2C uptake must be interpreted with caution. Saris and Niva (1994) found that

63

ruthenium red inhibited swelling of liver mitochondria induced by 20 mM Zn2C. Using atomic absorption, Brierley and Knight (1967) directly measured Zn2C import. However, these works differ substantially from ours in that they treated non-neural mitochondria with high added Zn2C of 20 mM or more in a sucrose buffer. It may be that Zn2C can enter mitochondria, but only when present at supraphysiological levels and in the absence of other permeant ions. Additionally, it should be noted that the uniporter is only vaguely understood, and the precise mechanism of ruthenium red remains unknown. It may be overstating then to confirm or exclude uniporter participation based exclusively on ruthenium red data. In summary, although our results show that ruthenium red partially protects from Zn2Cinduced loss of Djm, we have no other conclusive evidence supporting uniporter-mediated uptake of Zn2C in isolated brain mitochondria. 4.4. Zn2C and accumulation of ROS Mitochondria are presumed to be the major source of oxidative stress in cells. Our results show that Zn2C affects the rates of mitochondrial ROS accumulation under all three substrate conditions tested. Zn2C attenuated the relatively high rate of ROS accumulation resulting from succinate metabolism. Succinate-driven ROS accumulation in neural mitochondria results from reverse electron flow from complex II to complex I (Cino and Del Maestro, 1989). As shown by Votyakova and Reynolds (2001), and recently reconfirmed by Liu et al. (2002), accumulation of ROS under these conditions requires high Djm. Thus, the depolarizing effect of Zn2C accounts for Zn2C inhibition of succinate-supported ROS accumulation. In contrast, Zn2C increased ROS accumulation in mitochondria powered by GCM or G3P. While we are not certain, it is reasonable to assume that ROS accumulation under these conditions involves transfer from ubiquinone to complex III. However, generation of ROS at complex I cannot be ruled out. In summary, it is clear that ROS accumulation in isolated neural mitochondria can be augmented in the presence of pathophysiologically relevant [Zn2C].

64

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65

4.5. Zn2C and permeability transition Some reports suggest that Zn2C causes MPT. Wudarczyk et al. (1999) reported that micromolar concentrations of added Zn2C in the presence of various mitochondrial inhibitors caused MPT in liver mitochondria but Brown et al. (2000) saw no evidence for MPT in liver mitochondria exposed to Zn2C. A recent report claimed that added Zn2C-induced MPT in liver mitochondria (Bossy-Wetzel et al., 2004). Most relevant to the present report is the work of Jiang et al. (2001), who reported that brain mitochondria in sucrose-based medium underwent MPT in response to as little as 10 nM Zn2C. This resulted in the release of apotogenic factors and was inhibited by CsA. In the present study, the classic MPT inhibitor CsA did not protect mitochondria from Zn2C-induced loss of Djm. Perhaps more importantly, most Zn2C effects on Djm and O2 consumption were rapidly reversed by Zn2C chelation. If Zn2C caused MPT, one might expect these parameters to be altered in an irreversible manner, particularly if essential elements of the ETC such as cytochrome c were lost into the surrounding medium. Complete restoration of O2 consumption, as we observed upon addition of EGTA, is inconsistent with the substantial mitochondrial changes expected to accompany permeability transition. We are unable to explain why we find no evidence of Zn2C-induced MPT in neural mitochondria as reported by others. These disparate conclusions might result from different buffer conditions (sucrose vs. KCl based), alternatively MPT may for some reason occur only at the relatively low Zn2C concentrations employed by Jiang et al. who observed swelling at [Zn2C]Z10 nM. It is difficult however to understand how such precise and low [Zn2C] was achieved, as no system for controlling [Zn2C] was described. Zn2C has long been recognized as a ubiquitous and troublemaking contaminate, and one would expect that an unbuffered recording solution already contains nanomolar [Zn2C] (Frederickson, 1989).

5. Summary Zn2C inhibited O2 consumption and dissipated Djm under all physiological substrate conditions. Consequently, Zn2C may interdict all feasible

Fig. 7. Zn2C has numerous effects on mitochondrial function. Schematic shows route of electron delivery for each substrate condition. Zn2C binds at bc1 of complex III, probably at a site accessible from the intermembrane space, thereby impeding electron flow. However, inhibition of complex I cannot be excluded. Consequences of Zn2C inhibition include reduced O2 consumption, dissipation of Djm, and increased generation of ROS. Zn2C likely augments ROS production through inhibition of complex III. All of these effects are consistent with reduced mitochondrial ATP production. Zn2C uptake through the Ca2C uniporter remains ambiguous, and our data are inconsistent with Zn2C-induced MPT.

pathways of neural oxidative phosphorylation and concomitant ATP production (Fig. 7). Moreover, Zn2C-induced accumulation of mitochondrial ROS raises the possibility of a forward feeding mechanism in which Zn2C-induced-ROS accumulation could liberate more zinc from protein bound zinc stores (Maret and Vallee, 1998; Aizenman et al., 2000). These results suggest several routes by which zinc may cause neuronal death.

Acknowledgements We thank our colleagues for helpful discussion, Dr Gordon Rintoul for assistance with data analysis, and Dr Theresa Hastings for use of a polarographer. This work was supported by NIH grant NS34138 (IJR) and American Heart Association predoctoral training grant 9910111U (KED).

References Aizenman, E., Stout, A.K., Hartnett, K.A., Dineley, K.E., McLaughlin, B., Reynolds, I.J., 2000. Induction of neuronal apoptosis by thiol oxidation: putative role of intracellular zinc release. J. Neurochem. 75, 1878–1888.

K.E. Dineley et al. / Mitochondrion 5 (2005) 55–65 Akerman, K.E., Wikstrom, M.K., 1976. Safranine as a probe of the mitochondrial membrane potential. FEBS Lett. 68, 191–197. Berry, E.A., Zhang, Z., Bellamy, H.D., Huang, L., 2000. Crystallographic location of two Zn(2C)-binding sites in the avian cytochrome bc(1) complex. Biochim. Biophys. Acta 1459, 440–448. Bossy-Wetzel, E., Talantova, M.V., Lee, W.D., Scholzke, M.N., Harrop, A., Mathews, E., Gotz, T., Han, J., Ellisman, M.H., Perkins, G.A., Lipton, S.A., 2004. Crosstalk between nitric oxide and zinc pathways to neuronal cell death involving mitochondrial dysfunction and p38-activated KC channels. Neuron 41, 351–365. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. Brierley, G.P., Knight, V.A., 1967. Ion transport by heart mitochondria. X. The uptake and release of Zn2C and its relation to the energy-linked accumulation of magnesium. Biochemistry (Mosc) 6, 3892–3901. Brown, A.M., Kristal, B.S., Effron, M.S., Shestopalov, A.I., Ullucci, P.A., Sheu, K.F., Blass, J.P., Cooper, A.J., 2000. Zn2C inhibits alpha-ketoglutarate-stimulated mitochondrial respiration and the isolated alpha-ketoglutarate dehydrogenase complex. J. Biol. Chem. 275, 13441–13447. Canzoniero, L.M.T., Turetsky, D.M., Choi, D.W., 1999. Measurement of intracellular free zinc concentrations accompanying zinc-induced neuronal death. J. Neurosci. 19, 1–6. Cheng, C., Reynolds, I.J., 1998. Calcium-sensitive fluorescent dyes can report increases in intracellular free zinc concentration in cultured forebrain neurons. J. Neurochem. 71, 2401–2410. Cino, M., Del Maestro, R.F., 1989. Generation of hydrogen peroxide by brain mitochondria: the effect of reoxygenation following postdecapitative ischemia. Arch. Biochem. Biophys. 269, 623–638. Cuajungco, M.P., Lees, G.J., 1998. Nitric oxide generators produce accumulation of chelatable zinc in hippocampal neuronal perikarya. Brain Res. 799, 118–129. Dineley, K.E., Richards, L.L., Votyakova, T.V., Reynolds, I.J., 2001. Zn2C causes loss of membrane and alters production of reactive oxygen species in isolated brain mitochondria. Soc. Neurosci. Abstr. 27, 2311. Dineley, K.E., Malaiyandi, L.M., Reynolds, I.J., 2002. A reevaluation of neuronal zinc measurements: artifacts associated with high intracellular dye concentration. Mol. Pharmacol. 62, 618–627. Dineley, K.E., Votyakova, T.V., Reynolds, I.J., 2003. Zinc inhibition of cellular energy production: implications for mitochondria and neurodegeneration. J. Neurochem. 85, 563–570. Frederickson, C.J., 1989. Neurobiology of zinc and zinc-containing neurons. Int. Rev. Neurobiol. 31, 145–238. Jiang, D., Sullivan, P.G., Sensi, S.L., Steward, O., Weiss, J.H., 2001. Zn(2C) induces permeability transition pore opening and release of pro-apoptotic peptides from neuronal mitochondria. J. Biol. Chem. 276, 47524–47529. Link, T.A., von Jagow, G., 1995. Zinc ions inhibit the QP center of bovine heart mitochondrial bc1 complex by blocking a protonatable group. J. Biol. Chem. 270, 25001–25006.

65

Liu, Y., Fiskum, G., Schubert, D., 2002. Generation of reactive oxygen species by the mitochondrial electron transport chain. J. Neurochem. 80, 780–787. Lorusso, M., Cocco, T., Sardanelli, A.M., Minuto, M., Bonomi, F., Papa, S., 1991. Interaction of Zn2C with the bovine-heart mitochondrial bc1 complex. Eur. J. Biochem. 197, 555–561. Maret, W., Vallee, B.L., 1998. Thiolate ligands in metallothionein confer redox activity on zinc clusters. Proc. Natl Acad. Sci. USA 95, 3478–3482. Nicholls, D.G., Budd, S.L., 2000. Mitochondria and neuronal survival. Physiol. Rev. 80, 315–360. Nicholls, P., Malviya, A.N., 1968. Inhibition of nonphosphorylating electron transfer by zinc. The problem of delineating interaction sites. Biochemistry (Mosc) 7, 305–310. Noh, K.M., Kim, Y.H., Koh, J.Y., 1999. Mediation by membrane protein kinase C of zinc-induced oxidative neuronal injury in mouse cortical cultures. J. Neurochem. 72, 1609–1616. Outten, C.E., O’Halloran, T.V., 2001. Femtomolar sensitivity of metalloregulatory proteins controlling zinc homeostasis. Science 292, 2488–2492. Reed, K.C., Bygrave, F.L., 1974. The inhibition of mitochondrial calcium transport by lanthanides and ruthenium red. Biochem. J. 140, 143–155. Rosenthal, R.E., Hamud, F., Fiskum, G., Varghese, P.J., Sharpe, S., 1987. Cerebral ischemia and reperfusion: prevention of brain mitochondrial injury by lidoflazine. J. Cereb. Blood Flow Metab. 7, 752–758. Saris, N.E., Niva, K., 1994. Is Zn2C transported by the mitochondrial calcium uniporter?. FEBS Lett. 356, 195–198. Sensi, S.L., Canzoniero, L.M., Yu, S.P., Ying, H.S., Koh, J.Y., Kerchner, G.A., Choi, D.W., 1997. Measurement of intracellular free zinc in living cortical neurons: routes of entry. J. Neurosci. 17, 9554–9564. Sensi, S.L., Yin, H.Z., Carriedo, S.G., Rao, S.S., Weiss, J.H., 1999. Preferential Zn2C influx through Ca2C-permeable AMPA/kainate channels triggers prolonged mitochondrial superoxide production. Proc. Natl Acad. Sci. USA 96, 2414–2419. Skulachev, V.P., Chistyakov, V.V., Jasaitis, A.A., Smirnova, E.G., 1967. Inhibition of the respiratory chain by zinc ions. Biochem. Biophys. Res. Commun. 26, 1–6. Turrens, J.F., 1997. Superoxide production by the mitochondrial respiratory chain. Biosci. Rep. 17, 3–8. Votyakova, T.V., Reynolds, I.J., 2001. DeltaPsi(m)-Dependent and -independent production of reactive oxygen species by rat brain mitochondria. J. Neurochem. 79, 266–277. Weiss, J.H., Sensi, S.L., Koh, J.Y., 2000. Zn(2C): a novel ionic mediator of neural injury in brain disease. [erratum appears in Trends Pharmacol Sci. 2000 Dec;21(12):496]. Trends Pharmacol. Sci. 21, 395–401. Wudarczyk, J., Debska, G., Lenartowicz, E., 1999. Zinc as an inducer of the membrane permeability transition in rat liver mitochondria. Arch. Biochem. Biophys. 363, 1–8. Zamzami, N., Kroemer, G., 2001. The mitochondrion in apoptosis: how Pandora’s box opens. Nat. Rev. Mol. Cell Biol. 2, 67–71.