Superoxide production in mitochondria

In earlier studies [5,6] superoxide detection in mitochondria was equated to hydrogen peroxide formation. However, while it is quite possible that superoxide is a stoichiometric precursor of mitochondrial hydrogen peroxide, it is understandable that the level of hydrogen peroxide may be decreased due to the reactions with various mitochondrial oxidants. Moreover, superoxide level can be underestimated due to the reaction with mitochondrial MnSOD. Several authors [7,8] assumed that mitochondrial superoxide production may be estimated through cyanide-resistant respiration, which supposedly characterizes univalent dioxygen reduction. This method was applied for the measurement of superoxide production under in vitro normoxic and hyperoxic conditions, in spite of the finding [7] that cyanide-resistant respiration reflects also the oxidation of various substrates (lipids, amino acids, and nucleotides). Earlier,... 

As in the case of superoxide production by prooxidant enzymes described in Chapter 22, reliable data on mitochondrial superoxide production have been received by the use of lucigenin-amplified CL [12]. Lucigenin cation easily accumulates in mitochondria with the negative mitochondrial membrane potential and is able to penetrate the inner membrane of mitochondria. Therefore, this method permits to measure superoxide production by the whole mitochondria and mitochondrial superoxide production in cells. For example, Esterline and Trush [13] registered a significant lucigenin-amplified CL produced by the mitochondria of unstimulated rat alveolar macrophages that cannot be due to the activity of dormant NADPH oxidase. 

Despite a long-time studying of superoxide production by mitochondria, an important question is still debated does mitochondria produce superoxide under physiological conditions or superoxide release is always a characteristic of some pathophysiological disorders resulting in the damage of normal mitochondrial functions Uncertainties in this question arise due to the different results obtained with the use of respiratory inhibitors and different analytical methods. 

Now, we may consider in detail the mechanism of oxygen radical production by mitochondria. There are definite thermodynamic conditions, which regulate one-electron transfer from the electron carriers of mitochondrial respiratory chain to dioxygen these components must have the one-electron reduction potentials more negative than that of dioxygen Eq( 02 /02]) = —0.16 V. As the reduction potentials of components of respiratory chain are changed from 0.320 to +0.380 V, it is obvious that various sources of superoxide production may exist in mitochondria. As already noted earlier, the two main sources of superoxide are present in Complexes I and III of the respiratory chain in both of them, the role of ubiquinone seems to be dominant. Although superoxide may be formed by the one-electron oxidation of ubisemiquinone radical anion (Reaction (1)) [10,22] or even neutral semiquinone radical [9], the efficiency of these ways of superoxide formation in mitochondria is doubtful. 

Such a process is supposed to occur within the limits of Q-cycle mechanism (Figure 23.2). In accord with this scheme ubihydroquinone reduced dioxygen in Complex III, while superoxide producers in Complex I could be FMN or the FeS center [12]. Zhang et al. [24] also suggested that the Q-cycle mechanism is responsible for the superoxide production by the succinate-cytochrome c reductase in bovine heart mitochondria and that FAD of succinate dehydrogenase is another producer of superoxide. Young et al. [25] concluded that, in addition to Complex III, flavin-containing enzymes and FeS centers are also the sites of superoxide production in liver mitochondria. 

Calcium oxalate monohydrate responsible for the formation of most kidney stones significantly increased mitochondrial superoxide production in renal epithelial cells [42], Recombinant human interleukin IL-(3 induced oxygen radical generation in alveolar epithelial cells, which was suppressed by mitochondrial inhibitors 4 -hydroxy-3 -methoxyacetophe-none and diphenylene iodonium [43]. Espositio et al. [44] found that mitochondrial oxygen radical formation depended on the expression of adenine nucleotide translocator Anti. Correspondingly, mitochondria from skeletal muscle, heart, and brain from the Antl-deficient mice sharply increased the production of hydrogen peroxide. 

As a rule, oxygen radical overproduction in mitochondria is accompanied by peroxidation of mitochondrial lipids, glutathione depletion, and an increase in other parameters of oxidative stress. Thus, the enhancement of superoxide production in bovine heart submitochondrial particles by antimycin resulted in a decrease in the activity of cytochrome c oxidase through the peroxidation of cardiolipin [45]. Iron overload also induced lipid peroxidation and a decrease in mitochondrial membrane potential in rat liver mitochondria [46]. Sensi et al. [47] demonstrated that zinc influx induced mitochondrial superoxide production in postsynaptic neurons. 

Important studies were performed by Trush and coworkers [42], who showed the advantages of applying lucigenin-amplified CL for the measurement of superoxide production by mitochondria in unstimulated monocytes and macrophages as well as by isolated mitochondria [43,44]. Later on, these authors have shown that mitochondrial superoxide production measured by lucigenin-amplified CL increased in the liver of rats treated with the promoter of hepatocarcinogenesis ethinyl estradiol [45], in liver from obese mice [46], and in children with Down syndrome [47]. 

Han D, Antunes F, Canali R, Rettori D, Cadenas E (2003a) Voltage-dependent anion channels control the release of the superoxide anion from mitochondria to cytosol. J Biol Chem 278(8) 5557-5563 Han D, Canali R, Rettori D, Kaplowitz N (2003b) Effect of glutathione depletion on sites and topology of superoxide and hydrogen peroxide production in mitochondria. Mol Pharmacol 64 (5) 1136-1144... 

The functions of mtNOS in mitochondria have been studied (see Chapter 23). Ghafourifar et al. [177] found that the calcium-induced stimulation of mtNOS caused the release of cytochrome c from mitochondria and induced apoptosis. On the other hand, the same group of authors [178] showed that the production of NO by mtNOS and superoxide in mitochondria resulted in the formation of peroxynitrite and stimulated calcium release, indicating the existence of a feedback loop which prevents calcium overload in mitochondria. 

Controversial results of oxygen radical detection in mitochondria have been described in the literature. Owing to experimental difficulties many authors were obliged to work with sub-mitochondrial particles instead of the whole mitochondria. However, it is quite possible that oxygen radical production by submitochondrial particles may be artificially enhanced due to exposure to oxygen. On the other hand, some analytical methods of superoxide detection such as cytochrome c reduction cannot be used due to the direct reduction of cytochrome by mitochondrial components. 

Classic antioxidants, vitamin E, vitamin C, and others can suppress the activation of apoptosis. For example, ascorbic acid prevented cytochrome c release and caspase activation in human leukemia cells exposed to hydrogen peroxide [128], Pretreatment with A -acctylcystcinc, ascorbate, and vitamin E decreased homocysteine thiolactone-induced apoptosis in human promyelocytic leukemia HL-60 cells [129]. Resveratrol protected rat brain mitochondria from anoxia-reoxygenation damage by the inhibition of cytochrome c release and the reduction of superoxide production [130]. However, it should be mentioned that the proapoptotic effect of ascorbate, gallic acid, or epigallocatechin gallate has been shown in the same human promyelocytic leukemia cells [131]. 

---