CA2354743A1 - Mitochondrially targeted antioxidants - Google Patents

Abstract

A mitochondrially-targeted antioxidant compounds preferably of the formula (see formula I) wherein Z is a suitable anionic species and n is an integer from 1 to 25.

Description

MITOCHONDRIALLY TARGETED ANTIOXIDANTS
TECHNICAL FIELD
The invention relates to antioxidants having a lipophilic cationic group and to uses of these antioxidants, for example, as pharmaceuticals.
BACKGROUND OF THE INVENTION
Oxidative stress contributes to a number of human degenerative diseases associated with ageing, such as Parkinson's disease, and Alzheimer's disease, as well as to Huntington's Chorea, diabetes and Friedreich's Ataxia, and to non-specific damage that accumulates with aging. It also contributes to inflammation and ischaemic-reperfusion tissue injury in stroke and heart attack, and also during organ transplantation and ), surgery. To prevent the damage caused by oxidative stress a number of antioxidant therapies have been developed. However, most of these are not targeted within cells and are therefore less than optimally effective.
Mitochondria are intracellular organelles responsible for energy metabolism.
Consequently, mitochondria) defects are damaging, particularly to neural and xrluscle tissues which have high energy demands. They are also the major source of the free radicals and reactive oxygen species that cause oxidative stress inside most cells.
Therefore, the applicants believe delivering antioxidants selectively to mitochondria will be more effective than using non-targeted antioxidants. Accordingly, it is towards the provision of antioxidants which may be targeted to mitochondria that the present invention is directed.
Lipophilic cations may be accumulated in the mitochondria) matrix because of their positive charge (Rottenberg, ( 1979) Methods Enzymol, 55, 547-560; Chen, ( 1988) Annu Rev Cell Biol 4, 155-181). Such ions are accumulated provided they are sufficiently lipophilic to screen the positive charge or delocalise it over a large surface area, also provided that there is no active efflux pathway and the canon is not metabolised or immediately toxic to a cell.

Thiobutyl triphenylphosphonium bromide [TBTP] is disclosed in R J Burn and M P
Murphy, (1997), Archives ofBiochemi8stry and Biophysics, Vol. 339, No. 1, March 1, pp 33-39 as a probe selectively accumulated by mitochondria in living cells that binds selectively to mitochondria) proteins. They also disclose that TBTP does have some antioxidant activity provided endogenous glutathione is depleted. RA) Smith et al, Eur.
J. Biochem, 263, 709-716 ( 1999) "Selective targeting of an antioxidant to mitochondria"
discloses the content of our Canadian Patent Specification No. 2311318.
US Patent 3532&67discloses triphenyl phosphonium halide compositions for use as an antioxidant in polyamides. (J Chem. Soc. Perkin Trans I 1984) pp 709-21 discloses a certain triphenyl phosphoniurn bromide compound useful for the purpose of synthesis of an antioxidant. Tetrahedron Letters No. 23, pp 2145 - 2148 1979 discloses certain triphenylphosphonium bromides used in the synthesis of benzofurannes.
The focus of the invention is therefore on an approach by which it is possible to use the ability of mitochondria to concentrate compounds comprising triphenphosphonium cations carbon linked to certain quinol antioxidant moieties which target to the major source of free radicals and reactive oxygen species causing oxidative stress and which compounds have a capability it is believed to anchor within mitochondria.

SUMMARY OF THE INVENTION
In a first aspect the present invention is a mitochondrially-targeted antioxidant compound comprising or including a triphenlphosphonium cation linked by a C1 to C3o carbon chain (optionally including none, one or more of each of or either double or triple bonds, and optionally including one or more subsdtuents) to an antioxidant moiety being (1')m or OH
where m is an integer from 0 to 3, and Y is independently selected from alkoxy, thioalkyl, alkyl, haloalkyl, halo, amino, nitro and optionally substituted aryl.
Preferably the compound also includes an anion (eg; that of Br).
Preferably the C1 to C3o carbon chain is an alkyl chain of the formula -(CH2)n-where n is an integer of from 1 to 25 and preferably 2 to 25.
Preferably n is 5 or more.
Most preferably Y is independently selected from alkoxy and alkyl.
Preferably m is 2 or 3.
Preferably the compound is OH
CH3~
Z
(CH2)n I' OH
wherein Z is a suitable anionic species and n is ~an integer from 1 to 25.

Most preferably n is 10 and the compound is OH

CH O

In another aspect the invention is a method of therapy of prophylaxis of a mammalian patient who would benefit from reduced oxidative stress which comprises or includes the step of administering to the patient a mitochondrially-targeted antioxidant of the present invention.
In still a further aspect the invention is a method of reducing oxidative stress in a cell which comprises or includes the step of administering to the cell a mitochondrially targeted antioxidant of the present invention.
In yet another aspect the invention is a pharmaceutical antioxidant dosage unit or composition which includes a compound of the present invention.
DESCRIPTION OF DRAWINGS
In particular, a better understanding of the invention will be gained with reference to the accompanying drawings, in which:
Figure 1 is a graph which shows the uptake by isolated mitochondria of compound 1, a Vitamin E derivative (see Example 1) coupled by a carbon chain to a triphenylphosphonium moiety, a mitochondrially-targeted antioxidant according to our Canadian Patent Specification No. 2311318 (Canadian NPE of PCT/NZ98/00173).
Figure 2 is a graph which shows the accumulation of compound 1 by isolated mitochondria;
Figure 3 is a graph which shows a comparison of a compound 1 uptake with that of the triphenylphosphonium cation (TPMP);

Figure 4 is a graph which shows that compound 1 protects mitochondria against oxidative damage;
Figure 5 is a graph which compares compound 1 with vitamin E and the effect of uncoupler and other lipophilic cations;
Figure 6 is a graph which shows that compound 1 protects mitochondria) function from oxidative damage;
Figure ? is a graph which shows the effect of compound 1 on mitochondria) function;
Figure 8 is a graph which shows the uptake of compound 1 by cells;
Figure 9 is a graph which shows the energisation-sensitive uptake of compound 1 by cells;
Figure 10 is a graph which shows the effect of compound 1 on cell viability;
Figure 11 shows the UV-absorption spectrum of [10-(6'-ubiquinonyl)decyltriphenyl-phosphonium bromide] (herein referred to as "mitoquinone") and of the reduced form of the compound (10-(6'-ubiquinolyl)decyltriphenylphosphoniurn bromide]
(herein referred to as "mitoquinol");
Figures 12A to 12D show reactions of [ 10-(6'-ubiquinonyl)decyltriphenylphosphonium bromide) ("mitoquinone") and the reduced form of the compound ("mitoquinol") with mitochondria) membranes, both mitoquinone and mitoquinol being separately and together a compound of the present invention;
Figure 13 shows reactions of mitoquinol and mitoquinone with pentane-extracted mitochondria) membranes;
Figure 14 shows reduction of mitoquinone by intact mitochondria;
Figure 15 shows uptake of radiolabelled mitoquinol by energised rat liver mitochondria and its release on addition of the uncoupler FCCP;
Figure 16 shows the effect of mitoquinol on isolated rat liver mitochondria;

Figure 1? shows TPMP accumulation from oral administration to mice;
Figure 18 shows compound 1 accumulation from oral administration to mice; and Figure 19 shows uptake of mitoquinol by human osteosarcoma 143(3 cells.
DESCRIPTION OF THE INVENTION
As stated above, the focus of this invention is on the mitochondria) targeting of compounds, primarily for the purpose of therapy and/or prophylaxis to reduce oxidative stress.
Mitochondria have a substantial membrane potential of up to 180 mV across their inner membrane (negative inside). Because of this potential, membrane permeant, lipophilic cations accumulate several-hundred fold within the mitochondria) matrix.
The applicants have found that by covalently coupling lipophilic cations (preferably the lipophilic triphenylphosphonium cation) to an antioxidant the compound can be delivered to the mitochondria) matrix within intact cells. The antioxidant is then targeted to a primary production site of free radicals and reactive oxygen species within the cell, rather than being randomly dispersed. See our Canadian Patent Application No. 2311318 (Canadian NPE of PCT/NZ98/00173).
While it is generally preferred that the carbon chain is an alkyl chain (eg -(CI-i2)n-) (preferably C1-C2o, more preferably C1-C1;), carbon chains which optionally include none, one or more of each of or both double or triple bonds are also within the scope of the invention. Also included are carbon chains which include one or more substituents (such as hydroxyl, carboxylic acid or amide groups), and/or include one or more side chains or branches (for example, selected from unsubstituted or substituted alkyl, alkenyl or alkynyl groups).
In some particularly preferred embodiments, the linking group is an ethylene, propylene, butylene, pentylene or decylene group.
Preferred antioxidant compounds of the invention, can be readily prepared, for example, by the following reaction:

_7_ I \
//~ B r / -+P ~ ~ ~ Br \ P \
The general synthesis strategy is to heat a halogenated precursor, preferably a brominated or iodinated precursor (RBr or RI) in an appropriate solvent with 2-equivalents of triphenylphosphine under argon for several days. The phosphonium compound is then isolated as its bromide or iodide salt. To do this the solvent is removed, the product is then triturated repeatedly with diethyl ether until an off white solid remains. This is then dissolved in chloroform and precipitated with diethyl ether to remove the excess triphenylphasphine. This is repeated until the solid no longer dissolves in chloroform. At this point the product is recrystallised several times from methylene chloride/diethyl ether.
It will also be appreciated that the anion of the antioxidant compound thus prepared, which will be a halogen when this synthetic procedure is used, can readily be exchanged with another pharmaceutically or pharmacologically acceptable anion, if this is desirable or necessary, using ion exchange chromatography or other techniques known in the art.
The same general procedure can be used to make a wide range of mitochondrially targeted compounds with different antioxidant moieties R attached to the triphenylphosphonium cation.
In some preferred embodiments of the invention, the antioxidant compound is a quinol derivative of the formula II defined above. A particularly preferred quinol derivative of the invention is the compound mitoquinol as defined above. Another preferred compound of the invention is a compound of formula II in which (C)n is (CHZ)5, and the quinol moiety is the same as that of mitoquinol.

-$-Once prepared, the antioxidant compound of the invention, in any pharmaceutically appropriate form and optionally including pharmaceutically-acceptable carriers or additives, will be administered to the patient requiring therapy and/or prophylaxis.
Once administered (eg; orally, parenterally or otherwise - preferably orally) the compound will target the mitochondria within the cell.
The invention will now be described in more detail with reference to the following non-limiting examples.
EXAMPLES
Example 1 Experimental 1. Synthesis of a mitochondrially-targeted vitamin-E derivative (Compound I) The synthesis strategy for a mitochondrially-targeted vitamin-E derivative (compound 1) is as follows. The brominated precursor (compound 2) 2-(2-bromoethyl)-3,4-dihydro-6-hydroxy-2,5,7,8-tetramethyl-2H-1-benzopyran was synthesized by bromination of the corresponding alcohol as described by Grisar et al, ( 1995) (J Nled Chem 38, 2880-2886). The alcohol was synthesized by reduction of the corresponding carboxylic acid as described by Cohen et al., (1979) (J. AmerChem Soc 101, 6710-6716).
The carboxylic acid derivative was synthesized as described by Cohen et al., ( 1982) (Syn Commun 12, 57-65) from 2,6-dihydroxy-2,5,7,8-tetramethylchroman, synthesized as described by Scott et al., (1974) (J. Amer. Oil Chem. Soc. 101,6710-6716).

Br Compound 1 HC

Br Compound 2 For the synthesis of compound 1, 1g of compound 2 was added to 8 ml butanone containing 2.5 molar equivalents of triphenylphosphine and heated at 100°C in a sealed Kimax tube under argon for 7-8 days. The solvent was removed under vacuum at room temperature, the yellow oil triturated with diethyl ether until an off white solid remained. This was then dissolved in chloroform and precipitated with diethyl ether.
This was repeated until the solid was insoluble in chloroform and it was then recrystallised several times from methylene chloride/diethyl ether and dried under vacuum to give a white hygroscopic powder.
2. Mitochoadrial uptake of compound 1 To demonstrate that this targeting is effective for compounds our Canadian Patent Specification No. 2311318 and of the present invention, the exemplary vitamin E
compound 1 was tested in relation to both isolated mitochondria and isolated cells. To do this a [3H]-version of compound 1 was synthesized using (3H]-triphenylphosphine and the mitochondria) accumulation of compound 1 quantitated by scintillation counting (Fig. 1) (Burns et al., 1995, Arch Biochem Biophys 332,60-68; Burns and Murphy, 1997, Arch Biochem Biophys 339, 33-39). To do this rat liver mitochondria were incubated under conditions known to generate a mitochondria) membrane potential of about 180 mV (Burns et al., 1995; Burns and Murphy, 1997). Under these conditions compound 1 was rapidly (< 10 s) taken up into mitochondria with an accumulation ratio of about 6,000. This accumulation of compound 1 into mitochondria was blocked by addition of the uncoupler FCCP (carbonyl cyanide-p-trifluoromethoxyphenylhydrazone) which prevents mitochondria establishing a membrane potential (Figs. 1 and 2) (Burns et al., 1995). Therefore compound 1 is rapidly and selectively accumulated into mitochondria driven by the mitochondria) membrane potential and this accumulation results in a concentration of the compound within mitochondria several thousand fold higher than in the external medium.
This accumulation is rapidly (< 10 s) reversed by addition of the uncoupler FCCP to dissipate the mitochondria) membrane potential after accumulation of compound 1 within the mitochondria. Therefore the mitochondria) specific accumulation is solely due to the mitochondria) membrane potential and is not due to specific binding or covalent interaction.
The mitochondria) specific accumulation of compound 1 also occurs in intact cells.
This was measured as described by Burns and Murphy, 1997 and the accumulation was prevented by dissipating both the mitochondria) and plasma membrane potentials.
In addition, compound 1 was not accumulated by cells containing defective mitochondria, which consequently do not have a mitochondria) membrane potential.
Therefore the accumulation of compound 1 into cells is driven by the mitochondria) membrane potential.
The accumulation ratio was similar across a range of concentrations of compound 1 and the amount of compound 1 taken inside the mitochondria corresponds to an intramitochondrial concentration of 4-8 mM (Fig 2). This uptake was entirely due to the membrane potential and paralleled that of the simple triphenylphosphonium cation TPMP over a range of membrane potentials (Fig 3). From comparison of the uptake of TPMP and compound 1 at the same membrane potential we infer that within mitochondria about 84% of compound 1 is membrane-bound (cf. About 60% for the less hydrophobic compound TPMP).
Further details of the experimental procedures and results are given below.
Figure 1 shows the uptake of 10 ~M [3H] compound 1 by energised rat liver mitochondria (continuous line and filled symbols). The dotted line and open symbols show the effect of addition of 333 nM FCCP at 3 min. Incubation with FCCP from the start of the incubation led to the same uptake as for adding FCCP at 3 min (data not shown). Liver mitochondria were prepared from female Wistar rats by homogenisation followed by differential centrifugation in medium containing 250 mM sucrose, 10 mM
Tris-HCL (pH 7.4) and 1 mM EGTA and the protein concentration determined by the biuret assay using BSA as a standard. To measure [3H] compound 1 uptake mitochondria (2 mg protein/ml) were suspended at 25°C in 0.5 - 1 ml 110 mM KCl, 40 mM Hepes-KOH, pH 7.2, 0.1 mM EDTA supplemented with nigericin (1 ~g/ml), 10 mM
succinate, rotenone 1.33 ~g/ml and 60 nCi/ml [3H] compound 1 and 10 ~M
compound 1. After the incubation mitochondria were pelleted by centrifugation and the [3H) compound 1 in the supernatant and pellet quantitated by scintilation counting.
Figure 2 shows the mitochondrial accumulation ratios [(compound 1 /mg protein)/(compound 1/~1)] obtained following 3 min incubation of energised rat liver mitochondria with different concentrations of compound 1 (filled bars) and the effect of 333 nM FCCP on these (open bars). The dotted line and open circles show compound 1 uptake by mitochondria, corrected for FCCP-insensitive binding. To measure [3H) compound 1 accumulation ratio mitochondria (2 mg protein/ml) were suspended at 25°C in 0.5 - 1 ml 110 mM KCI, 40 mM Hepes-KOH, pH 7.2, 0.1 mM EDTA
supplemented with nigericin (1 ~g/ml), lOmM succinate, rotenone 1.33 ~g/ml and 60 nCi/ml [3H] compound 1 and 1-50 ~.M compound 1. After the incubation mitochondria were pelleted by centrifugation and the [3H] compound 1 in the supernatant and pellet quantitated by scintillation counting.
Figure 3 shows a comparison of compound 1 uptake with that of TPMP at a range of mitochondria) membrane potentials. Energised rat liver mitochondria were incubated for 3 min with 10 ~uM compound 1 and 1 ~M TPMP and different membrane potentials established with 0-8 mM malonate or 333 nM FCCP. The accumulation ratios of parallel incubations with either 60 nCi/ml [3H] compound 1 or 50 nCi/ml [3H) TPMP
were determined, and the accumulation ratio for compound 1 is plotted relative to that of TPMP at the same membrane potential (slope = 2.472, y intercept = 319, r =
0.97).
Mitochondria (2 mg protein/ml) were suspended at 25°C in 0.5-1 ml 110 mM KCl, 40 mM Hepes-KOH, pH 7.2, 0.1 mM EDTA supplemented with nigericin (lug/ml), 10 mM
succinate, rotenone 1.33 ~g/ml.
3. Anti-oxidant efficacy of compound 1 The compounds of the present invention are highly effective against oxidative stress.
To demonstrate this, compound 1 was further tested using rat brain homogenates. The rat brain homogenates were incubated with or without various concentrations of the test compounds (compound l; native Vitamin E (a-tocopherol), bromobutyl triphenylphosphonium bromide, Trolox (a water soluble form of Vitamin E) and compound 2, ie2-(2-bromoethyl)-3,4-dihydro-2,5,7,8-tetramethyl-2H-1-benzopyran-ol, the precursor of compound 1 ("Brom Vit E")) and the oxidative damage occurring over the incubation was quantitated using the TBARS assay (Stocks et al., 1974, Clin Sci Mol Med 47,215-222). From this the concentration of compound required to inhibit oxidative damage by 50% was determined. In this system 210 nM compound 1 inhibited oxidative stress by 50% while the corresponding value for native Vitamin E
was 36 nM. The value for bromobutyltriphenylphosphonium bromide, which contains the triphenylphosphonium moiety but not the antioxidant Vitamin E moiety was 47 ~ M.
These data show that compound 1 is an extremely effective antioxidant, within an order of magnitude as effective as Vitamin E. Comparison with bramobutyltriphenylphosphonium bromide shows that the antioxidant capability is due to the Vitamin E function and not to the phosphonium salt. Further details of the experimental procedures and results are set out below.
The ICSO values for inhibition of lipid peroxidation were determined in rat brain homogenates, and are means + , SEM or range of determinations on 2-3 brain preparations. Octan-1-ol/PBS partition coefficients are means ~ SEM for three independent determinations. N.D. not determined. Partition coefficients were determined by mixing 200 ~M of the compound in 2 ml water-saturated octanol-1-0l with 2 ml octanol-saturated-PBS at room temperature for 1 h, then the two layers were separated by brief centrifugation and their concentrations determined spectrophotometrically from standard curves prepared in PBS or octanol. To measure antioxidant efficacy four rat brains were hornogenised in I5 ml 40 mM
potassium phosphate (pH 7.4), 140 mM NaCl at 4°C, particulate matter was pelleted (1,000 x g at 4°C for 15 min) and washed once and the combined supernatants stored frozen.
Aliquots were rapidly thawed and 5 mg protein suspended in 800 ~l PBS
containing antioxidant or ethanol carrier and incubated at 37°C for 30 min.
Thiobarbituric acid reactive species (TBARS) were quantitated at 532 nm by adding 200 ~1 conc.
HC104 and 200 ~l 1% thiobarbituric acid to the incubation, heating at 100°C for 15 min and then cooling and clarification by centrifugation (10,000 x g for 2 min). The results are shown in Table 1 below.

Table 1. Partition coefficients and antioxidant efficacy of compound 1 and related compounds Compound ICso for inhibition Octanol:PBS partition of lipid coefficient peroxidation (nM) Compound 1 210 + 58 7.37 + 1.56 Bromo Vit E 45 + 26 33.1 + 4.4 a-Tocopherol 36 22 27.4 1.0 Trolox 18500 + 5900 N.D.

BrBTP 47000 + 13000 3.83 + 0.22 When mitochondria were exposed to oxidative stress corripound 1 protected them against oxidative damage, measured by lipid peroxidation and protein carbonyl formation (Fig 4). This antioxidant protection was prevented by incubating mitochondria with the uncoupler FCCP to prevent uptake of compound l, and lipophilic cations alone did not protect mitochondria (Fig 5). Most importantly, the uptake of compound 1 protected mitochondria) function, measured by the ability to generate a membrane potential, far more effectively than Vitamin E itself (Fig 6). This shows that the accumulation of compound 1 into mitochondria selectively protects their function from oxidative damage. In addition, we showed that compound 1 is not damaging to mitochondria at the concentrations that afford protection (Fig 7).
The next step was to determine whether compound 1 was accumulated by intact cells.
Compound 1 was rapidly accumulated by intact 143B cells, and the amount accumulated was greater than that by p° cells derived from 143B cells.
This is important because the p° cells lack mitochondria) DNA and consequently have far lower mitochondria) membrane potential than the 143B cells, but are identical in every other way, including plasma membrane potential, cell volume and protein content (Fig 8); this suggests that most of the compound 1 within cells is mitochondria). A
proportion of this uptake of compound 1 into cells was inhibited by blocking the plasma and mitochondria) membrane potentials (Fig. 9). This energisation-sensitive uptake corresponds to an intra mitochondria) concentration of compound 1 of about 2-4 mM, which is sufficient to protect mitochondria from oxidative damage. These concentrations of compound 1 are not toxic to cells (Fig. 10).
Further details of the experimental procedures and results are discussed below.

Figure 4 shows the protection of mitochondria against oxidative damage by compound 1. Mitochondria were exposed to oxidative stress by incubation with iron/ascorbate and the effect of compound 1 on oxidative damage assessed by measuring TBARS
(filled bars) and protein carbonyls (open bars). Rat liver mitochondria (10 mg protein) were incubated at 25°C in a shaking water bath in 2 ml medium containing 100 mM KCl, 10 mM Tris, pH 7.7, supplemented with rotenone (1.33 ~ug/ml), 10 mM succinate, 500 ~M
ascorbate and other additions. After preincubation for 5 min, 100 ~uM FeS04 was added and 45-55 min later duplicate samples were removed and assayed for TBARS
or protein carbonyls.
Figure 5 shows a comparison of compound 1 with vitamin E and the effect of uncoupler and other lipophilic cations. Energised rat liver mitochondria were exposed to tert-butylhydroperoxide and the effect of compound 1 (filled bars), a-tocopherol (open bars), compound 1 + 333 nM FCCP (stippled bars) or the simple lipophilic cation bromobutyl triphenylphosphonium (cross hatched bars) on TBARS formation determined. Rat liver mitochondria (4 mg protein) were incubated in 2 rnl medium containing 120 mM
KCl, 10 mM Hepes-HCl pH 7.2, 1 mM EGTA at 37°C in a shaking water bath for 5 min with various additions, then tert butyl hydroperoxide (5 mM) was added, and the mitochondria incubated for a further 45 min and then TSARS determined.
Figure 6 shows how compound 1 protects mitochondrial function from oxidative damage. Energised rat liver mitochondria were incubated with iron/ascorbate with no additions (stippled bars), 5 ~M compound 1 (filled bars), 5 ~M a-tocopherol (open bars) or 5 ~M TPMP (cross hatched bars), and then isolated and the membrane potential generated by respiratory substrates measured relative to control incubations in the absence of iron/ascorbate. Rat liver mitochondria were incubated at 25°C in a shaking water bath in 2 ml medium containing 100 mM KCl, 10 mM Tris, pH 7.7, supplemented with rotenone (1.33 ug/ml), 10 mM succinate, 500 ~M ascorbate and other additions.
After preincubation for 5 min, 100 ~M FeS04 was added and after 30 min the incubation was diluted with 6 ml ice-cold STE 250 mM sucrose, 10 mM Tris-HCL
(pH

7.4) and 1 mM EGTA, pelleted by centrifugation (5 min at 5,000 x g) and the pellet resuspended in 200 u1 STE and 20 ~l (= 1 mg protein) suspended in 1 ml 110 mM
KCl, mM HEPES, 0.1 M EDTA pH 7.2 containing 1 ~M TPMP and 50 nCi/ml [3H] TPMP
35 either 10 mM glutamate and malate, 10 mM succinate and rotenone, or 5 mM
ascorbate/ 100 ~uM TMPD with rotenone and myxothiazol (2 ~g/ml), incubated at 25°C
for 3 min then pelleted and the membrane potential determined as above and compared with an incubation that had not been exposed to oxidative stress.

Figure ~ shows the effect of compound 1 on the membrane potential (filled bars) and respiration rate of coupled (open bars), phosphorylating (stippled bars) and uncoupled mitochondria {cross hatched bars), as a percentage of values in the absence of compound 1. The effect of various concentrations of compound 1 on the membrane potential of isolated mitochondria was determined from the distribution of [3H] TPMP
by incubating rat liver mitochondria (2 mg protein/ml) in 0.5 ml medium as above containing 1 ~M TPMP and 50 nCi/ml [3H] TPMP at 25°C for 3 min. After the incubation mitochondria were pelleted by centrifugation and the [3H] TPMP in the supernatant and pellet quantitated by scintilation counting and the membrane potential calculated assuming a volume of 0.5 ~rl/mg proteins and that 60% of intramitochondrial TPMP is membrane bound. To measure the effect of compound 1 on coupled, phosphorylating and uncoupled respiration rates, mitochondria (2 mg protein/ml) were suspended in 120 mM KCI, 10 mM Hepes-HCl pH 7.2, 1 mM EGTA, 10 rnM K Pi in a 3 ml Clark oxygen electrode then respiratory substrate, ADP
(200~M) and FCCP (333 nM) were added sequentially to the electrode and respiration rates measured.
Figure 8 shows the uptake of compound 1 by cells. Here 106 143B cells (closed symbols) or p° cells (open symbols) were incubated with 1 ~M [3H]
compound l and the compound 1 accumulation ratio determined. Human osteosarcoma 143B cells and a derived p° cell line lacking rnitochondrial DNA were cultured in DMEM/
10 % FCS (foetal calf serum) supplemented with uridine and pyruvate under an atmosphere of 5%
COZ/95% air at 37°C, grown to confluence and harvested for experiments by treatment with trypsin. To measure [3H] compound 1 accumulation cells (106) were incubated in 1 ml HEPES-buffered DMEM. At the end of the incubation, cells were pelleted by centrifugation, the cell pellet and the supernatant prepared for scintillation counting and the accumulation ratio [compound 1 / mg protein) / (compound 1 / ~ul)]
calculated.
Figure 9 shows the amount of compound 1 taken up by 106 143B cells over 1 h incubation, corrected for inhibitor-insensitive binding. Human osteosarcoma cells were incubated in 1 ml HEPES-buffered DMEM with 1-50 ~M compound 1 supplemented with 6-60 nCi/ml [3H] compound 1. To determine the energistration-dependent uptake, parallel incubations with 12.5 uM oligomycin, 20 pM FCCP, 10 ~M
myxathiazol, 100 nM valinomycin and 1mM ouabain were carried out. At the end of the incubation, cells were pelleted by centrifugation and prepared for scintillation counting and the energisation-sensitive uptake determined.

Figure 10 shows the effect of compound 1 on cell viability. Here, confluent 143B cells in 24 well tissue culture dishes were incubated with various concentrations of compound 1 for 24 h and cell viability measured by lactate dehydrogenase release.
Figure 11 shows the UV-absorption spectrum of [10-(6'-ubiquinonyl)decyltriphenyl-phosphonium bromide] (herein referred to as "mitoquinone") and of the reduced form of the compound [10-(6'-ubiquinolyl)decyltriphenylphosphonium bromide] (herein referred to as "mitoquinol"), both separately or together being a compound of the present invention;
Figures 12A to 12D show reactions of [10-(6'-ubiquinonyl)decyltriphenylphosphonium bromide] ("mitoquinone") and the reduced form of the compound ("mitoquinol") with mitochondria) membranes. Beef heart mitochondria) membranes (20 ~g/ml) were suspended in 50 mM sodium phosphate, pH 7.2 at 20°C. In panel A
rotenone and antimycin were present and for the t = 0 scan, then succinate (5 mM) was added and scans repeated at 5 minute intervals as indicated. In panel B A27swas monitored in the presence of rotenone and antimycin and then mitoquinone (50 ~uM) was added, followed by succinate (5 mM) and malonate (20 mM) where indicated. In Panel C rotenone, ferricytochrome c (50 ~M) and malonate (20 mM) were present, A2~swas monitored and mitoquinol (50 ~M) and myxathiazol (10 ~M) were added where indicated. In panel D
Assowas monitored and the experiment in Panel C was repeated in the presence of KCN.
Addition of myxathiazol inhibited this rate by about 60 - 70%. There was no reaction between mitoquinone and succinate or NADH in the absence of mitochondria) membranes, however mixing 50 ~uM mitoquinone, but not mitoquinol, with 50 ~M
ferricytochrome c led to some reduction of Asso;
Figure 13 shows reactions of mitoquinol and mitoquinone with pentane-extracted mitochondria) membranes. Pentane extracted beef heart mitochondria ( 100~g protein/ml) were suspended in 50 mM sodium phosphate, pH 7.2 at 20°C.
In Panel A
NADH (125 ~M) was added and A34o was monitored and ubiquinone-1 (UQ-1; 50 ~M) added where indicated. This was repeated in Panel b, except that mitoubiquinone (50 ~M) was added. In Panel C pentane extracted mitochondria were incubated with mitoquinone (50 ~uM), A2,swas monitored and succinate (5 mM) and malonate (20 mM) added where indicated. In Panel D pentane-extracted mitochondria were incubated with NADH (125 ~M), ferricytochrome c (50 ~M) and Asso was monitored and mitoquinone (50 ~M) was added where indicated. Addition of myxathiazol inhibited the rate of reduction by about 60 - 70 %;
Figure 14 shows reduction of mitoquinone by intact mitochondria. Rat liver mitochondria (100 ~g/ml) were incubated in 120 mM KCl, 10 mM HEPES, 1 mM EGTA, pH 7.2 at 20°C and A275 monitored. In panel A rotenone and succinate (5 mM) were present and mitoquinone (50 ~M) was added where indicated. This experiment was repeated in the presence of malonate (20 mM) or FCCP (333 nM). In panel B
glutamate and malate (5 mM of each) were present from the start and and mitoquinone (50 ~uM) was added where indicated. This experiment was repeated in the presence of FCCP or with rotenone and FCCP. Addition of TPMP (50 ~M) instead of mitoquinone did not lead to changes in A27s;
Figure 15 shows uptake of radiolabelled mitoquinol by energised rat liver mitochondria and its release on addition of the uncoupler FCCP;
Figure 16 shows the effect of mitoquinol on isolated rat liver mitochondria.
In A rat liver mitochondria energised with succinate were incubated with various concentrations of mitoquinol and the membrane potential determined as a percentage of control incubations. In B the respiration rate of succinate energised mitochondria under state 4 (black), state 3 (white) and uncoupled (stippled) conditions, as a percentage of control incubations.
Distribution of TPMP ovithia mice:
Pairs of mice were supplied with drinking water supplemented with 500~M TPMP
spiked with tritiated TPMP for various times. The mice were then killed and the amount of TPMP in each organ quantitated by homogenisation followed by scintillation counting. This showed that substantial amounts of TPMP did accumulate in the major organs, in particular there was substantial uptake by the brain and heart.
This corresponds to about 5-10 ~M in the brain and about 15-20 ~M in the heart. See Figure 17.
Distribution of Mit Vit E within mice:
Pairs of mice were supplied with drinking water supplemented with 500 ~uM Mit Vit E
spiked with tritiated Mit Vit E for various times. The mice were then killed and the amount of Mit Vit E in each organ quantitated by homogenisation followed by scintillation counting. Uptake is expressed as nmol Mit Vie E/g wet weight of the organ, except for blood where the uptake is in nmol/ml. This led to substantial uptake into their major organs. The uptake was less than for TPMP but was still substantial in the brain and heart. This corresponds to about 1 ~M in the brain and about ~M in the heart, which are sufficient to offer protection from oxidative stress.
Example 2 Synthesis of [10-(6'-ubiquinolyl)decyltriphenplphosphonium bromide] (herein referred to as "mitoquinol") Synthesis of precursors To synthesise 11-bromoundecanoic peroxide 11-bromoundecanoic acid (4.00 g, 15.1 mmol) and SOCl2 (1.6 mL, 21.5 mmol) were heated, with stirring, at 90°C
for 15 min.
ExcessSOCl2 was removed by distillation under reduced pressure (15 mm Hg, 90°C) and the residue (IR; 1799 crri') was dissolved in diethyl ether (20 mL) and the solution cooled to 0°C. Hydrogen peroxide (30%, 1.8 mL) was added, followed by dropwise addition of pyridine (1.4 mL) over 45 min. Diethyl ether (10 mL) was added and the mixture was stirred for 1 h at room temperature then diluted with diethyl ether ( 150 mL) and washed with H20 (2 x 70 mL), 1.2 M HCI (2 x 70 mL), H20 (70 mL), 0.5 M
NaHC03 (2 x 70 mL) and H20 (70 mL). The organic phase was dried over MgS04 and the solvent removed at room temperature under reduced pressure, giving a white solid (3.51 g). IR (nujol mull) 1810, 1782.
6-( 10-bromodecyl)ubiquinone was synthesised by mixing crude material above (3.51 g, 12.5 mmol max), (ubiquinoneo, 1.31 g, 7.19 mmol, Aldrich) and acetic acid (60 mL) and stirring the mixture for 20 h at 100°C. The mixture was diluted with diethyl ether (600 mL) and washed with H20 (2 x 400 mL), 1 M HCI (2 x 450 mL), 0.50 M NaHC03 (2 x 450 mL) and H2O (2 x 400 mL). The organic phase was dried over MgS04. The solvent was removed under reduced pressure, giving a reddish solid (4.31 g).
Column chromatography of the crude solid on silica gel (packed in CHZC12) and elution with CH2Cl2 gave the product as a red oil (809 mg, 28%) and unreacted ubiquinone as a red solid (300 mg, 1.6 mmol, 13%). TLC: Rf (CH2CI2, diethyl ether 20:1) 0.46; IR
(neat) 2928, 2854, 1650, 161 l, 1456, 1288; ~m~ (ethanol): 278 nm; 1H NMR (299.9 MHz) 3.99 (s, 6H, 2 x-OCH3), 3.41 (t, J= 6.8 Hz, 3H, -CH2-Br), 2.45 (t, J= 7.7 Hz, 2H, ubquinone-CHZ-), 2.02, (s, 3H, -CH3). 1.89 (quin, J= 7.4 Hz, 3H, -CHZ -CHZ -Br), 1.42-1.28 (m, 20H, -(CHZ)7-); 13C NMR (125.7 MHz) 184.7 (carbonyl), 184.2 (carbonyl), 144.3 (2C, ring), 143.1 (ring), 138.7 (ring), 61.2 (2 x-OCH3), 34.0 (-CH2-); 32.8 (-CHZ-), 29.8 (-CH2-), 29.4 (2 x -CH2-), 29.3 (-CH2-), 28.7 (2 x -CH2-), 28.2 (-CH2-), 26.4 (-CH2-), 11.9 (-CH3). Anal.

Calcd. For C19H2902Br:C, 56.86; H, 7.28; Found: C, 56.49, H, 7.34; LREI mass spectrum: calcd. For C19H2902Br 400/402; Found 400/402.
To form the quinol, 6-(10-bromodecyl)-ubiquinol, Na.BH4 (295 mg, 7.80 mmol) was added to a solution of the quinone (649 mg, 1.62 mmol) in methanol (6 mL) and stirred under argon for 10 min. Excess NaBH4 was quenched with 5% HCI (2 mL) and the mixture diluted with diethyl ether (40 mL). The organic phase was washed with 1.2 M
HCl (40 mL) and saturated NaCI (2 x 40 mL), and dried over MgS04. The solvent was removed under reduced pressure, giving a yellow oily solid (541 mg, 83%). 1H
NMR
(299.9 MHz) 5.31 (s, 1H, -OH), 5.26 (s, 1H, -OH), 3.89 (s, 6H, 2 x-OCH3), 3.41 (t, J=
6.8 Hz, 2H, -CH2 -Br), 2.59 (t, J = 7.7 Hz, 2H ubquinol-CH2-), 2.15 (s, 3H, CH3) 1.85 (quin, J =7.4 Hz, 2H, -CHZ -CH2 -Br), 1.44-1.21 (m, 19H, -CH2)~-).
Synthesis of 10-(6'-ubiquinolyl)decyltriphenylphosphonium bromide ('mitoquinol') To synthesise 10-(6~-ubiquinolyl)decyltriphenylphosphonium bromide. To a 15 mL
Kimax tube were added 6-( 10-bromodecyl)ubiquinol (541 mg, 1.34 mmol), PPH3 (387 mg, 1.48 mmol), ethanol (95%, 2.5 mL) and a stirnng bar. The tube was purged with argon, sealed and the mixture stirred in the dark for 88 h at 85°C. The solvent was removed under reduced pressure, giving an oily orange residue. The residue was dissolved in CH2C12 (2 mL) followed by addition of pentane (20 mL). The resultant suspension was refluxed for 5 min at 50°C and the supernatant decanted.
The residue was dissolved in CH2C12 (2 mL) followed by addition of diethyl either (20 mL).
The resultant suspension was refluxed for 5 min at 40°C and the supernatant decanted.
The CH2C12/diethyl ether reflux was repeated twice more. Residual solvent was removed under reduced pressure, giving crude product as a cream solid (507 mg). 1H
NMR (299.9 MHz) 7.9-7.6 (m, 20H, -P+ Ph3), 3.89 (s, 6H, 2 x-OCH3), 3.91-3.77 (m, 2H, -CH2-P+Ph3), 2.57 (t, J= 7.8 Hz, 2H ubquinol-CH2-), 2.14 (s, 3H, CH3), 1.6-1.2 (m, 23H, -(CH2)$-). 31P NMR (121.4 MHz) 25.1.
The crude product (200 mg) was oxidized to 10-(6'-ubiquinonyl)decyltriphenylphosphonium bromide (the oxidised form) by stirring in CDC13 under an oxygen atmosphere for 13 days. The oxidation was monitored by'H
NMR and was complete after 13 days. The solvent was removed under reduced pressure and the resultant residue dissolved in CH2CI2 (5 mL). Excess diethyl ether (15 mL) was added and the resultant suspension stirred for 5 min. The supernatant was decanted and the CHzCI2/diethyl ether precipitation repeated twice more.
Residual solvent was removed under reduced pressure, giving crude product as a brown sticky solid (173 mg).
The quinone was reduced to the quinol by taking a mixture of crude quinor~e and quinol (73 mg, ca. 3:1 by 1 H NMR) in methanol ( 1 mL) was added NaBH4 (21 mg, 0.55 mmol). The mixture was stirred slowly under an argon atmosphere for 10 min.
Excess NaBH4 was quenched with 5% HBr (0.2 mL) and the mixture extracted with CHZCI2. The organic extract was washed with H20 (3 x 5 mL). Solvent was removed under reduced pressure, giving a mixture of quinone and quinol (ca 1:5 by'H
NMR) as a pale yellow solid (55 mg).For routine preparation of the quinol form the ethanolic solution, dissolve in 5 vols of water, (= 1 ml) add a pinch of NaBH4 leave on ice in the dark for 5 min, then extract 3 x 0.5 ml dichloromethane, Wash with water/HCl etc blow off in nitrogen; dissolve in same vol of etoh and take spectrum and store at -80 under argon. Yield about 70 - 80%. Oxidises rapidly in air so should be prepared fresh. vortex with 1 ml 2M NaCl. Collect the upper organic phase and evaporate to dryness under a stream of N2 and dissolve in 1 ml ethanol acidified to pH 2.
Synthesis of [ 3HJ-10-(6'-ubiquinonyl)decyltriphenylphosphonium bromide To a Kimax tube was added 6-(10-bromodecyl)ubiquinol ( 6.3 mg; 15.6 ~mol) triphenylphosphine (4.09 mg; ' 15.6 ~mol) and 100 ~1 ethanol containing [3H]
triphenylphosphine (74 ~Ci custom synthesis by Moravek Biochemicals, Brea, CA, USA, Spec Ac 1 Ci/mmol) and 150 ~ul ethanol added. The mixture was stirred in the dark under argon for 55h at 80°C. Then it was cooled and precipitated by adition of 5 ml diethyl ether. The orange solid was dissolved in few drops of dichloromethane and then precipitated with diethyl ether and the solid was washed (x4) with ~ 2 ml diethyl ether. Then dissolved in ethanol to give a stock solution of 404 ~M which was stored at -20°C. The UV absorption spectrum and TLC were identical to those of the unlabelled 10-(6'-ubiquinonyl)decyltriphenylphosphonium bromide and the specific activity of the stock solution was 2.6 mCi/mmol.
Distributioa of Mitoquinol within Mice:
Mice supplied with mitoquinol orally were subject to dosage related distribution of mitoquinol into the organs similarly to TPMF and Mit Vit E (compound 1).

With the reduced water solubility of mitoquinol with respect to compound 1, an oil suspension delivery into the mice was adopted in preferenceto a supplemented drinking water delivery. Nevertheless uptake into the organs and flushing therefrom was demonstrated as was a ability with ongoing dosage of maintaining organ levels of the mitoquinol.
Extinction coe~cients Stock solutions of the quinone in ethanol were stored at -80°C in the dark and their concentrations confirmed by 31P nmr. The compound was converted to the fully oxidised form by incubation in basic 95% ethanol over an hour on ice or by incubation with beef heart mitochondria) membrane at room temperature, either procedure leading to the same extinction coefficient of 10,400 M-1 cm 1 at the local maximum of 275 nrn, with shoulders at 263 and 268 nm corresponding to the absorption maxima of the triphenylphosphonium moiety (Smith et al, Eur. J. Biochem., 263, 709-716, 1999;
Burns et al; Archives of Biochemistry and Biophysics, 322, 60-68, 1995) and a broad shoulder at 290 nm due to the quinol (Crane et al, Meth. Enzymol., 18C, 137-165, 1971). Reduction by addition of NaBH4 gave the spectrum of the quinol which had the expected peak at 290 nm with an extinction coefficient of 1800 M-1 cm-1 and the extinction coefficient for at 268nm was 3,000 M-' cm-1 the same as that for the phosphonium moiety alone (Burns, 1995 above). The extinction coefficient of 10,400 M-1 cm-1 at 275nm was lower than that for other quinones which have values of 14,600 M-1 cm-i in ethanol (Crane, 1971 above) and 12,250 M-1 cm-1 in aqueous buffer (Cabririi et al, Arch. Biochem Biaphys, 208, 11-19, 1981). While the absorbance of the quninone was about 10% lower in buffer than in ethanol, the discrepancy was not due to an interaction between the phosphonium and the quinone as the absorbance of the precursor quinone before linking to the phosphonium and that of the simple phosphonium methyltriphenylphosphonium were additive when 50 uM of each were mixed together in either ethanol or aqueous buffer. The DEoX _ rea was 7,000 M-lcrri 1 The spectrum of fully oxidised mitoquinone (50 ~M) in 50 mM sodium phosphate, pH
7.2 is shown in Figure 11. Addition of NaBH4 gave the fully reduced compound, mitoquinol. The UV absorption spectrum of the reduced (quinol) and oxidised (quinone) mitoquinone/ol are shown in Figure 11. To determine whether the mitochondria) respiratory chain could also oxidise or reduce the compound mitoquinone was incubated with beef heart mitochondria) membranes (Figure 12). In panel A the spectrum of fully oxidised mitoquinone in the presence of antimycin inhibited membranes is shown (t = 0; Fig 12A). Addition of succinate led to the gradual reduction of the mitoquinol as measured by repeating the measurement every five minutes and showing that the peak at 275 nm gradually disappeared, the presence of antimycin prevented the oxidation of the quinol by mitochondria) complex III. Succinate did not lead to the complete reduction of mitoquinone to mitoquinol, as can be seen by comparing the complete reduction brought about by borohydride (Fig 11), instead it reduced about 23 % of the added ubiquinone (Fig 12A). This is presumably due to equilibration of the quinol/quinone couple with the succinate/fumarate couple (Em Q=
40 mV at pH 7, Em Suc = 30 mV), hence this proportion corresponds to an Eh of about +8 mV.
The reduction of mitoquinone can be followed continuously at A2,5 nm (Fig 12B). On addition to rotenone inhibited mitochondria) membranes the small amount of mitoquinol remaining was oxidised leading to a slight increase in A275, but on addition of the Complex II substrate succinate mitoquinone was rapidly reduced and this reduction was blocked by malonate, an inhibitor of Complex II (Fig 12B). The rate of reduction of mitoquinone was 51 ~ 9.9 nmol /min/mg protein, which compares with the rate of reduction of cytochrome c by succinate in the presence of KCN of nmol/min/mg. Allowing for the 2 electrons required for mitoquinone reduction compared with 1 for cytochrome c the rate of electron flux into the mitoquinone pool is of similar order to the electron flux through the respiratory chain.
To determine whether mitoquinol was oxidised by Complex III of the respiratory chain, mitoquinol was added to beef heart membranes which had been inhibited with rotenone and malonate (Fig 12C). The mitoquinol was oxidised rapidly by membranes at an initial rate of about 89 ~ 9 nmol mitoquinol/min/mg protein (mean of 2 +/-range) and this oxidation was blocked by myxathiazol an inhibitor of complex III (Fig 12C). To confirm that these electrons were being passed on to cytochrome c, mitoquinol was then added to membrane supplemented with ferricytochrome c and the rate of reduction of cytochrome c monitored (Fig 12D). Addition of mitoquinol led to reduction of cytochrome c at an initial rate of about 93 +/- 13 nmol/min/mg (mean +/-range).
This rate was largely blocked by myxathiazol, although a small amount of cytochrome c reduction (about 30 - 40%) was not blocked by myxathiazol.
Mitoquinone/ol may be picking up and donating electrons directly from the active sites of the respiratory complexes, or it could be equilibrating with the endogenous mitochondrial ubiquinone pool. To address this question the endogenous ubiquinone pool was removed from beef heart mitochondria by pentane extraction. In the absence of endogenous ubiquinone as an electron acceptor the pentane extracted beef heart mitochondria could not oxidise added NADH, but addition of ubiquinone-l, a ubiquinone analogue that can pick up electrons from the active site of complex I, the oxidation of NADH is partially restored (Fig 13A). Similarly, addition of mitoquinone also restored NADH oxidation indicating that mitoquinone can pick up electrons from the complex I active site (Fig 13B). Succinate could also donate electrons to mitoquinone in pentane extracted beef heart mitochondrial in a malonate sensitive manner, suggesting that mitoquinone could also pick up electrons from the active site of Complex II (Fig 13C). Finally, the effect of the quinone on the flux of electrons to cytochrome c was detemined and it was shown that there was no NADH-ferricytochrome c activity until mitoquinone was added (Fig 13D), and this was partially inhibited by myxathizol (60 - 70 %).
The next step was to see if mitoquinone also accepted electrons within intact mitochondria (Fig 14). When mitoquinone was added to intact energised mitochondria it was rapidly reduced (Fig 14A). In the presence of the uncoupler FCCP to dissipate the membrane potential the rate was decreased about 2-3 fold, presumably due to the prevention of the uptake of the compound in to the mitochondria (Fig 14A). The complex II inhibitor malonate also decreased the rate of reduction of mitoquinone (Fig 14A). Use of the NADH-linked substrates glutamate/malate also led to the rapid reduction of mitoquinone by intact mitochondria which again was decreased by addition of the uncoupler FCCP (Fig 14B). The Complex I inhibitor rotenone also decreased the rate of reduction of mitoquinone (Fig 14B).
The next step was to see if mitoquinol was accumulated by energised mitochondria.
To do this a tritiated version of the compound was made, incubated with energised mitochondria and the amount taken up into the mitochondria determined. It can be seen that the compound is accumulated rapidly and that this accumulation is reversed by addition of the uncoupler FCCP (Fig 15).
The next assays were to determine the toxicity of these compounds to mitochondria and cells. To determine the toxicity to isolated mitochondria the effect on membrane potential and respiration rate were measured (Fig 16). It can be seen from Figure 16 that 10 ~M mitoquinol had little effect on mitochondrial function and at 25 ~M
and above there was some uncoupling and inhibition of respiration.

F"agure 19 shows the uptake of mitoquinol by Human osteosarcoma 143B cells (5 x 106) were incubated with 5 ~M (3H]-mitoquinol (filled squares) or with 4 ~M FCCP
(open triangle) or with a range of inhibitors (open circles).
INDUSTRIAL APPLICATION
The compounds of the invention have application in selective antioxidant therapies for human patients to prevent mitochondria) damage. This can be to prevent the elevated mitochondria) oxidative stress associated with particular diseases, such as Parkinson's disease, diabetes or diseases associated with mitochondria) DNA mutations.
They could also be used in conjunction with cell transplant therapies for neurodegenerative diseases, to increase the survival rate of implanted cells.
In addition, these compounds could be used as prophylactics to protect organs during transplantation, or ameliorate the ischaemia-reperfusion injury that occurs during surgery. The compounds of the invention could also be used to reduce cell damage following stroke and heart attack or be given prophylactically to premature babies, which are susceptible to brain ischemia. The methods of the invention have a major advantage over current antioxidant therapies - they will enable antioxidants to accumulate selectively in mitochondria, the part of the cell under greatest oxidative stress. This will greatly increase the efficacy of antioxidant therapies.
Related lipophilic canons are being trialed as potential anticancer drugs and are known to be relatively non-toxic to whole animals, therefore these mitochondrially-targeted antioxidants are unlikely to have harmful side effects.
Any appropriate delivery technique can be utilised ranging from orally, parenterally, etc. in the mammalian patient to in vitro into tissues or cells to be transplanted into the mammalian patient.
Those persons skilled in the art will appreciate that the above description is provided by way of example only, and that different lipophilic cation/antioxidant combinations can be employed without departing from the scope of the invention.

Claims (13)

1. A mitochondrially-targeted antioxidant compound comprising or including a triphenlphosphonium cation linked by a C1 to C30 carbon chain (optionally including none,one or more of each of or either double or triple bonds, and optionally including one or more substituents) to an antioxidant moiety being where m is an integer from 0 to 3, and Y is independently selected from alkoxy, thioalkyl, alkyl, haloalkyl, halo, amino, nitro and optionally substituted aryl.

2. A compound as claimed in claim 1 which also includes an anion.

3. A compound of claim 1 or claim 2 wherein the anion is that of Br.

4. A compound of any one of the preceeding claims wherein the C1 to C30 carbon chain is an alkyl chain of the formula -(CH2)n- where n is an integer of from 1 to 25.

5. A compound of claim 4 wherein n is 5 or more.

6. A compound as claimed in any one of the preceding claims wherein Y is independently selected from alkoxy and alkyl.

7. A compound as claimed in claim 6 wherein m is 2 or 3.

8. A compound as claimed in any one of the preceding claims wherein the compound is wherein Z is a suitable anionic species and n is an integer from 1 to 25.

9. A compound as claimed in claim 8 wherein n is 10 and the compound is

10. A compound of claim 9 wherein Z is Br.

11. A method of therapy of prophylaxis of a patient who would benefit from reduced oxidative stress which comprises or includes the step of administering to the patient a mitochondrially-targeted antioxidant as defined in any one of the preceding claims.

12. A method of reducing oxidative stress in a cell which comprises or includes the step of administering to the cell a mitochondrially targeted antioxidant as defined in any one of claims 1 to 10.

13. A pharmaceutical antioxidant dosage unit or composition which includes a compound of any one of claims 1 to 10.