Mitochondria-targeted antioxidants in the treatment of disease.

Mitochondrial oxidative damage is thought to contribute to a wide range of human diseases; therefore, the development of approaches to decrease this damage may have therapeutic potential. Mitochondria-targeted antioxidants that selectively block mitochondrial oxidative damage and prevent some types of cell death have been developed. These compounds contain antioxidant moieties, such as ubiquinone, tocopherol, or nitroxide, that are targeted to mitochondria by covalent attachment to a lipophilic triphenylphosphonium cation. Because of the large mitochondrial membrane potential, the cations are accumulated within the mitochondria inside cells. There, the conjugated antioxidant moiety protects mitochondria from oxidative damage. Here, we outline some of the work done to date on these compounds and how they may be developed as therapies.

Targeting lipoic acid to mitochondria: synthesis and characterization of a triphenylphosphonium-conjugated alpha-lipoyl derivative.

Lipoic acid (LA) is a widely used antioxidant that protects mitochondria from oxidative damage in vivo. Much of this protection is thought to be due to the reduction of LA to dihydrolipoic acid (LAH(2)). This reduction is catalyzed in vivo by thioredoxin, thioredoxin reductase (TrxR), and lipoamide dehydrogenase. We hypothesized that specifically targeting LA to mitochondria, the site of most cellular reactive oxygen species production, would make it a more effective antioxidant. To do this, we made a novel molecule, MitoLipoic acid, by attaching lipoic acid to the lipophilic triphenylphosphonium cation. MitoL was accumulated rapidly within mitochondria several-hundred fold driven by the membrane potential. MitoL was reduced to the active antioxidant dihydroMitoLipoic acid by thioredoxin and by lipoamide dehydrogenase but not by TrxR. In isolated mitochondria or cells MitoL was only slightly reduced (5-10%), while, in contrast, LA was extensively reduced. This difference was largely due to the reaction of LA with TrxR, which did not occur for MitoL. Furthermore, in cells MitoL was quantitatively converted to an S-methylated product. As a consequence of its lack of reduction, MitoL was not protective for mitochondria or cells against a range of oxidative stresses. These results suggest that the protective action of LA in vivo may require its reduction to LAH(2) and that this reduction is largely mediated by TrxR.

Mitochondrial targeting of quinones: therapeutic implications.

Mitochondrial oxidative damage contributes to a range of degenerative diseases. Ubiquinones have been shown to protect mitochondria from oxidative damage, but only a small proportion of externally administered ubiquinone is taken up by mitochondria. Conjugation of the lipophilic triphenylphosphonium cation to a ubiquinone moiety has produced a compound, MitoQ, which accumulates selectively into mitochondria. MitoQ passes easily through all biological membranes and, because of its positive charge, is accumulated several hundred-fold within mitochondria driven by the mitochondrial membrane potential. MitoQ protects mitochondria against oxidative damage in vitro and following oral delivery, and may therefore form the basis for mitochondria-protective therapies.

Interaction of the mitochondria-targeted antioxidant MitoQ with phospholipid bilayers and ubiquinone oxidoreductases.

MitoQ(10) is a ubiquinone that accumulates within mitochondria driven by a conjugated lipophilic triphenylphosphonium cation (TPP(+)). Once there, MitoQ(10) is reduced to its active ubiquinol form, which has been used to prevent mitochondrial oxidative damage and to infer the involvement of reactive oxygen species in signaling pathways. Here we show MitoQ(10) is effectively reduced by complex II, but is a poor substrate for complex I, complex III, and electron-transferring flavoprotein (ETF):quinone oxidoreductase (ETF-QOR). This differential reactivity could be explained if the bulky TPP(+) moiety sterically hindered access of the ubiquinone group to enzyme active sites with a long, narrow access channel. Using a combination of molecular modeling and an uncharged analog of MitoQ(10) with similar sterics (tritylQ(10)), we infer that the interaction of MitoQ(10) with complex I and ETF-QOR, but not complex III, is inhibited by its bulky TPP(+) moiety. To explain its lack of reactivity with complex III we show that the TPP(+) moiety of MitoQ(10) is ineffective at quenching pyrene fluorophors deeply buried within phospholipid bilayers and thus is positioned near the membrane surface. This superficial position of the TPP(+) moiety, as well as the low solubility of MitoQ(10) in non-polar organic solvents, suggests that the concentration of the entire MitoQ(10) molecule in the membrane core is very limited. As overlaying MitoQ(10) onto the structure of complex III indicates that MitoQ(10) cannot react with complex III without its TPP(+) moiety entering the low dielectric of the membrane core, we conclude that the TPP(+) moiety does anchor the tethered ubiquinol group out of reach of the active site(s) of complex III, thus explaining its slow oxidation. In contrast the ubiquinone moiety of MitoQ(10) is able to quench fluorophors deep within the membrane core, indicating a high concentration of the ubiquinone moiety within the membrane and explaining its good anti-oxidant efficacy. These findings will facilitate the rational design of future mitochondria-targeted molecules.

Fine-tuning the hydrophobicity of a mitochondria-targeted antioxidant.

The mitochondria-targeted antioxidant MitoQ comprises a ubiquinol moiety covalently attached through an aliphatic carbon chain to the lipophilic triphenylphosphonium cation. This cation drives the membrane potential-dependent accumulation of MitoQ into mitochondria, enabling the ubiquinol antioxidant to prevent mitochondrial oxidative damage far more effectively than untargeted antioxidants. We sought to fine-tune the hydrophobicity of MitoQ so as to control the extent of its membrane binding and penetration into the phospholipid bilayer, and thereby regulate its partitioning between the membrane and aqueous phases within mitochondria and cells. To do this, MitoQ variants with 3, 5, 10 and 15 carbon aliphatic chains were synthesised. These molecules had a wide range of hydrophobicities with octan-1-ol/phosphate buffered saline partition coefficients from 2.8 to 20000. All MitoQ variants were accumulated into mitochondria driven by the membrane potential, but their binding to phospholipid bilayers varied from negligible for MitoQ3 to essentially total for MitoQ15. Despite the span of hydrophobicites, all MitoQ variants were effective antioxidants. Therefore, it is possible to fine-tune the degree of membrane association of MitoQ and other mitochondria targeted compounds, without losing antioxidant efficacy. This indicates how the uptake and distribution of mitochondria-targeted compounds within mitochondria and cells can be controlled, thereby facilitating investigations of mitochondrial oxidative damage.