Alloxan-induced mitochondrial permeability transition triggered by calcium, thiol oxidation, and matrix ATP.

In addition to their critical function in energy metabolism, mitochondria contain a permeability transition pore, which is regulated by adenine nucleotides. We investigated conditions required for ATP to induce a permeability transition in mammalian mitochondria. Mitochondrial swelling associated with mitochondria permeability transition (MPT) was initiated by adding succinate to a rat liver mitochondrial suspension containing alloxan, a diabetogenic agent. If alloxan was added immediately with or 5 min after adding succinate, MPT was strikingly decreased. MPT induced by alloxan was inhibited by EGTA and several agents causing thiol oxidation, suggesting that alloxan leads to permeability transition through a mechanism dependent on Ca(2+) uptake and sulfhydryl oxidation. Antimycin A and cyanide, inhibitors of electron transfer, carbonyl cyanide m-chlorophenylhydrazone, and oligomycin all inhibited MPT. During incubation with succinate, alloxan depleted ATP in mitochondria after an initial transient increase. However, in a mitochondrial suspension containing EGTA, ATP significantly increased in the presence of alloxan to a level greater than that of the control. These results suggest the involvement of energized transport of Ca(2+) in the MPT initiation. Addition of exogenous ATP, however, did not trigger MPT in the presence of alloxan and had no effect on MPT induced by alloxan. We conclude that alloxan-induced MPT requires mitochondrial energization, oxidation of protein thiols, and matrix ATP to promote energized uptake of Ca(2+).

Mitochondrial permeability transition (MPT) 1 is associated with an increase in the permeability of the mitochondrial inner membrane to solutes with a molecular mass of 1.5 kDa or lower and can disturb mitochondrial functions in several ways (1)(2)(3). It is generally accepted that MPT is due to the opening of a so-called permeability transition pore (PTP), which accompanies a loss of transmembrane potential (⌬), a release of Ca 2ϩ and other cations, and mitochondrial swelling, causing ATP synthesis to stop in mitochondria (2)(3)(4)(5)(6). The PTP is thought to have a role in cellular Ca 2ϩ homeostasis, in import of mitochondrial protein, in thermal regulation in mammalian mitochondria, and as a common mediator of cell death (7,8).
We focused on the role of adenine nucleotides in the modulation of MPT. The sensitivity of MPT to [Ca 2ϩ ] is greatly increased by ATP depletion (9). Matrix ADP is an important modulator of PTP opening and acts by decreasing the sensitivity of the calcium trigger site to Ca 2ϩ (10). The opening of PTP may be caused by well known membrane constituents, including the adenine nucleotide translocator (ANT), porin molecules, and the complex forming the peripheral benzodiazepine receptor (11,12). Two ADP binding sites may exist, one with a high affinity with the ANT that is blocked by the inhibitor atractyloside and the other site with a lower affinity (5,13). These findings suggest that adenine nucleotides may be negative regulators of MPT.
Mitochondrial damage seems to occur in the early phase of cell death (14), and evidence supports the idea that MPT is involved in apoptosis (6,11,(15)(16)(17)(18). Intracellular levels of ATP and mitochondrial dysfunction determine the way a cell dies with a difference between apoptosis and necrosis; ATP is required for apoptosis (19 -21). These findings suggest that ATP is involved in the apoptotic process that includes the induction of MPT. So far no clear evidence exists that ATP is involved in the initiation of MPT in mammalian mitochondria. This led us to explore the conditions that are necessary for ATP to induce the MPT.
In this study, the permeability transition of rat liver mitochondria progressed markedly by adding respiratory substrates such as succinate in the presence of alloxan. Alloxan, a diabetogenic agent, causes changes in respiration, a decrease in the concentration of adenine nucleotides, Ca 2ϩ release, and loss of ⌬ in mammalian mitochondria (22,23), and it has thiol oxidant activity similar to some MPT inducers (1)(2)(3). The focus of this study was to characterize the involvement of an increase in the ATP level, Ca 2ϩ uptake, and membrane thiol oxidation in alloxan-induced MPT initiated by adding succinate to rat liver mitochondria.

MATERIALS AND METHODS
Liver mitochondria were prepared daily from male Wistar rats weighing about 200 g that were fasted overnight as described previously (24). The permeability transition of mitochondria resulting from permeation of sucrose into the mitochondrial matrix was measured by recording the decrease in absorbance at 540 nm. The standard experimental conditions were as follows. Mitochondria (1 mg of protein/ml) were equilibrated in a total volume of 3 ml of Chelex 100-treated medium consisting of 0.1 M rotenone, 0.25 M sucrose, and 10 mM Tris-HCl at pH 7.4 and 37°C for 5 min. The suspension was then preincubated with 1 mM alloxan for 5 min. MPT was initiated by adding 5 mM succinate to this suspension. Various inhibitors, except for EGTA as indicated in the legend of Fig. 2, were included at the beginning of the preincubation period before adding alloxan. The effect of the various compounds on MPT was represented by changes in the maximal rate or extent of swelling as calculated from the change in absorbance at 10 min after adding succinate.
The amount of ATP in the mitochondria was measured using the luciferin-firefly luciferase method (25). Briefly, mitochondria (1 mg of * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‡ To whom correspondence should be addressed. Tel.: 81-134-62-1892; Fax: 81-134-62-5161; E-mail: ks51@hokuyakudai.ac.jp. 1 The abbreviations used are: MPT, mitochondrial permeability transition; ANT, adenine nucleotide translocator; CCCP, carbonyl cyanide m-chlorophenylhydrazone; ⌬, mitochondrial inner membrane potential; PTP, permeability transition pore. protein/ml) were preincubated in the presence or absence of alloxan, EGTA, or both at 37°C and then incubated with 5 mM succinate for the indicated time. Twenty l of the mitochondrial suspension was immediately mixed with 280 l of HEPES buffer containing 0.02% sodium azide and 200 l of the releasing solution. The amount of ATP was measured using an ATP assay kit. All operations for the assay ended after 90 s. To decrease the concentration of adenine nucleotides, mitochondria were incubated with P i as described previously (26,27).
The amount of thiol in the mitochondrial membrane was measured using 5,5Ј-dithio-bis(2-nitrobenzoic acid) at 412 nm as described previously (28). The experimental conditions were the same as those described for the MPT determination except that the incubation time was 10 min, and 50 g of protein/ml of mitochondria was used.
Alloxan was obtained from Wako Co. (Osaka, Japan) and was dissolved in Chelex 100-treated medium that had been purged with N 2 gas for 10 min. Oligomycin, carbonyl cyanide m-chlorophenylhydrazone (CCCP), rotenone, and antimycin A were obtained from Sigma Chemical Co. and were dissolved in ethanol. The ATP assay kit was from Roche Molecular Biochemicals. All other chemicals used in this study were of the highest grade available from commercial suppliers.
Data are expressed as means Ϯ S.E. and were statistically analyzed using the Student's t test for paired data. p Ͻ 0.05 was considered statistically significant.

RESULTS
Induction of Permeability Transition-After isolated rat liver mitochondria in the presence of rotenone were preincubated with succinate for 5 min, MPT was induced by adding Ca 2ϩ or tert-butyl hydroperoxide plus Ca 2ϩ (Fig. 1A). When alloxan was added 5 min after adding succinate, only a small MPT was observed. The features of MPT induced by these inducers agreed with other studies (3,4,22,29). In mitochondria that had previously been incubated with 1 mM alloxan for 5-15 min, an enhanced MPT was clearly observed after adding succinate (Fig. 1B). However, adding alloxan together with succinate to the mitochondrial suspension resulted in little MPT. Increases in alloxan concentrations up to 2 mM increased the rate and  extent of MPT initiated by adding succinate (Fig. 1C). The MPT was significantly inhibited by 1 M cyclosporin A (data not shown), suggesting that the absorbance changes indicated the "opening" of the PTP. Increases in succinate up to 5 mM in the presence of a constant concentration of alloxan induced the MPT in a concentration-dependent manner (data not shown). Because MPT in the presence of alloxan proceeded rapidly and markedly after adding succinate, subsequent experiments were made by first preincubating mitochondria with alloxan for 5 min in the presence of rotenone before adding succinate. Involvement of Ca 2ϩ and Membrane Protein Thiol-Ca 2ϩ is the most important factor for MPT, but the effects are modulated by a wide variety of physiological and chemical factors. Also, it has been shown that alloxan influences intramitochondrial [Ca 2ϩ ] (30). Fig. 2 shows that the MPT induced by alloxan was inhibited strongly by EGTA in a concentration-dependent manner ( Fig. 2A). EGTA added after the initiation of MPT prevented further swelling but did not reverse the progression of swelling (Fig. 2B). Mitochondria were preincubated with 0.05 mM EGTA to eliminate free Ca 2ϩ present in the mitochondrial preparation and then were preincubated with or without alloxan and various concentrations of Ca 2ϩ (Fig. 2C). Added Ca 2ϩ concentrations of more than 10 M in the presence of alloxan or more than 100 M in the absence of alloxan were sufficient to lead to MPT. These results suggest that Ca 2ϩ is necessary for the induction of the MPT by alloxan.
Alloxan reacts rapidly with protein thiol groups (23, 31) and some thiol oxidants and thiol cross-linking reagents can induce MPT (1,6,32). We therefore investigated whether membraneprotein thiol groups were involved in the MPT induced by alloxan. N-Ethylmaleimide and 4,4Ј-diisothiocyanatostilbene-2,2Ј-disulfonic acid strongly and diamide partially prevented MPT induced by alloxan (Table I). The thiol oxidants used in this study did not induce MPT in the absence of exogenous Ca 2ϩ . When the mitochondria were preincubated with alloxan, the amount of membrane thiol decreased in mitochondria in-cubated with or without malate/glutamate or succinate, indicating that alloxan directly consumes thiols of the mitochondrial membrane (Table II). These results suggest that the oxidation of membrane thiol protein is involved in the induction of MPT by alloxan. In the absence of alloxan, the amount of membrane thiol only slightly decreased after incubation with succinate but significantly increased after incubation with malate/glutamate. Adding malate/glutamate in place of succinate to a mitochondrial suspension clearly initiated MPT in the presence of alloxan, but the extent of MPT initiated by malate/ glutamate was significantly decreased compared with the extent of MPT initiated by succinate.
ATP Production Involved in Permeability Transition-To evaluate the role of succinate in the induction of MPT, we examined the effects of inhibitors of the electron transfer chain and of ATP synthesis on the MPT. Antimycin A, an inhibitor of ubiquinol-cytochrome c reductase (complex III), and KCN, an inhibitor of cytochrome oxidase reductase (complex IV), strongly inhibited the induction of MPT (Table III), implying that succinate functions as an electron donor in MPT induction. Also, CCCP, an uncoupler, and oligomycin, an inhibitor of F 0 F 1 -ATPase, significantly inhibited MPT. The MPT initiated by malate/glutamate was inhibited by rotenone, an inhibitor of NADH-ubiquinone oxidoreductase (complex I) as well as by inhibitors of complexes III and IV (data not shown). In the absence of alloxan, none of the test compounds had an effect on the permeability of mitochondria. These results suggest that ATP produced by the metabolism of respiratory substances in mitochondria is involved in the induction of MPT. Fig. 3 shows changes in the amount of ATP in mitochondria preincubated with alloxan to show the involvement of ATP. When the mitochondria were incubated with succinate in the presence or absence of EGTA, the level of ATP increased from 1 to 5 min after the start of incubation. In mitochondria preincubated with alloxan, the level of ATP transiently increased in the 1st min after adding succinate, reaching a value that was similar to that of control mitochondria, but the ATP level decreased until all the ATP was lost at 5 min after adding succinate. In the presence of EGTA, the ATP level in mitochondria preincubated with alloxan significantly increased compared with control mitochondria without EGTA and alloxan at 1 and 5 min after adding succinate, suggesting a link between the behavior of Ca 2ϩ and the consumption of ATP. Alloxan and EGTA had no effects on the production of ATP in the absence of succinate or on the light intensity of the luminescent reaction in the presence or absence of a sample. Under these conditions, CCCP and oligomycin inhibited the production of ATP after adding succinate in the presence or absence of alloxan (data not shown).
We next examined the effect of P i on the MPT. Pretreatment with P i decreased the rate of MPT induced by alloxan in a concentration-dependent manner (Fig. 4). When alloxan or succinate was omitted from the mixture, the range of P i concentrations used in this study did not induce MPT without added exogenous Ca 2ϩ .
Table IV (A) shows that neither exogenous ATP nor ADP was able to trigger MPT in mitochondria in the presence of alloxan. Adding exogenous ATP or atractyloside, an inhibitor of the ANT, had no effect on alloxan-induced MPT initiated by adding succinate, whereas exogenous ADP significantly inhibited the MPT (Table IV (B)). Unexpectedly, atractyloside significantly reduced the inhibition of MPT by exogenous ADP. These results suggest that ATP produced in mitochondria is involved in the induction of MPT by alloxan, and external ADP, but not ATP, inhibits the MPT.

DISCUSSION
In this study we showed that in mitochondria preincubated with alloxan, a diabetogenic agent, the membrane permeability transition is initiated by adding succinate by a reaction that can be inhibited by inhibitors of complex II, III, and IV in the electron transport chain. An uncoupler, CCCP, also inhibited the MPT. These results indicate that the induction of the MPT by alloxan is related to a transmembrane proton gradient generated by electron flow through the mitochondrial electron transfer system to molecular oxygen. Also, the induction of the MPT was clearly inhibited by oligomycin. However, adding exogenous ATP had no effect on the mitochondria without adding succinate in the presence of alloxan or on the MPT induced by alloxan. These results suggest that the ATP pro-duced would be involved in inducing MPT in the matrix or at a matrix site of the inner membrane.
Adenine nucleotides are often used as negative regulators of MPT in mammalian mitochondria initiated by some inducers (2,10,33). In this study, the external ADP significantly inhibited the MPT, which was reduced by atractyloside, an agent that can stabilize the conformation of ANT. Atractyloside itself had no effect on the MPT by alloxan. ANT catalyzes the exchange between ATP and ADP because adenine nucleotides cannot diffuse across the inner membrane. ADP enters the matrix only if ATP exists in the matrix. Therefore, we assume that external ADP is exchanged for endogenous ATP through the action of ANT, and this may result in a decrease in the ATP level in the mitochondrial matrix and inhibition of MPT induced by alloxan.
Lowering the matrix concentration of ADP may result in a decrease of the ADP-mediated inhibition of the MPT (10,26). If the decrease in the amount of ADP induces the MPT by alloxan, pretreatment with P i may increase MPT induced by alloxan by decreasing the matrix concentration of adenine nucleotides. Pretreatment with 5 mM P i for 10 min decreased the amount of ATP and ADP in mitochondria to about 20 and 40%, respectively, of the control (data not shown). Other studies showed similar results (26,35). However, the rate of MPT induced by alloxan in mitochondria pretreated with P i was lower than the MPT rate in control mitochondria. These results suggest that the induction of MPT by alloxan depends on an increase in the concentration of matrix ATP rather than a decrease in the concentration of ADP. Alloxan added before the respiratory substrates caused a slight reduction in state 3 respiration and a collapse of ⌬. However, alloxan added after the substrates had no effect on the respiration. 2 How alloxan causes an increase in the ATP level is not clear.
Adding EGTA before or after alloxan inhibited the MPT. Alloxan decreased the concentration of Ca 2ϩ needed to induce MPT. In the presence of EGTA, the increase in the ATP level was pronounced in mitochondria preincubated with alloxan compared with control mitochondria at 1 and 5 min after adding succinate. In contrast, a depletion of the ATP level occurred in mitochondria preincubated with alloxan in the absence of EGTA at 5 min. These results suggest that there is a link between the behavior of Ca 2ϩ and the consumption of ATP in the process of MPT. Transfer of Ca 2ϩ across the inner membrane is mediated by an energy-requiring process (36, 37).  Therefore, it seems reasonable that the ATP produced during electron transfer increases matrix [Ca 2ϩ ], which then initiates MPT. Alloxan, which is a mild thiol oxidant, directly consumes mitochondrial membrane thiols. The extent of MPT initiated by malate/glutamate was lower than the MPT initiated by succinate; malate/glutamate increased the membrane thiol content as found in other studies (38,39). It is thought that the thiols generated by metabolism of nicotinamide adenine dinucleotide reacts with alloxan, resulting in partial prevention against oxidation of the critical thiols involved in MPT. Several studies have shown that oxidation of membrane thiol groups is involved in MPT triggered by inducers such as diamide, arsenite, tert-butyl hydroperoxide, and 1-hydroxyethyl radicals (2, 24, 34, 40 -42). We showed that thiol oxidants had an effect on MPT induced by alloxan because thiol compounds such as dithiothreitol and N-acetylcysteine directly reacted with alloxan and reduced alloxan concentrations in the medium. Thiol oxidants inhibited MPT, but the degree of inhibition varied widely among the oxidants. The reactions of alloxan with specific thiols may be needed to induce MPT; this is in agreement with the conclusion of another study (28). Further investigation is needed to clarify this point.
The results of this study provide compelling evidence that a sequential addition of respiratory substances is needed to induce MPT by alloxan and indicate the existence of MPT in mammalian mitochondria resulting from the energized transport of Ca 2ϩ and oxidation of protein thiols. To the best of our knowledge, our results are the first to show that matrix ATP favors the pore opening in mammalian mitochondria. This finding supports the idea that MPT is important in the overall mechanism of apoptosis especially because apoptosis requires intracellular ATP (6, 11, 14, 18 -21). The association of MPT with cell injury mechanisms and the identification of ATP as an effective inducer of MPT should generate a high level of interest in this aspect of mitochondrial research.

Koichi Sakurai, Mika Katoh and Yukio Fujimoto
Thiol Oxidation, and Matrix ATP