WO2003035078A1

WO2003035078A1 - Combined use of catalytic antioxidants and glutathion or lipoic acid

Info

Publication number: WO2003035078A1
Application number: PCT/GB2002/004744
Authority: WO
Inventors: Peter Jeremy Cragg; Flavia Fucassi; Michael Harvey Lefevre Green
Original assignee: University Of Brighton
Priority date: 2001-10-22
Filing date: 2002-10-22
Publication date: 2003-05-01
Also published as: GB0125357D0

Abstract

A method for improving the safety and efficacy of catalytic antioxidant salen-metal complexes, such as salenMn(III) complexes, for possible clinical applications comprises combining them with a compound that enhances intracellular thiol levels, such as glutathione or α-lipoic acid. Hybrid compounds that comprise a catalytic salen-metal complex combined with a compound that enhances intracellular thiol levels are also provided, together with pharmaceutical compositions which comprise a catalytic antioxidant salen-metal complex or derivative thereof and a compound that enhances intracellular thiol levels.

Description

COMBINED USE OF CATALYTIC ANTIOXIDANTS AND GLUTATHION OR LIPOIC ACID.

This invention relates to improvements in the safety and efficacy of small molecule catalytic antioxidants, such as salen-manganese (III) complexes and derivatives, for potential clinical applications by combining them with a compound that enhances intracellular thiol levels. Hybrid molecules that combine a reducible disulphide or thiol and a catalytic moiety, using either covalent or ionic bonding, are also described.

US 5,403,834, US 5,696,109, US 5,827,880, US 5,834,509 and US 6,046,188 describe salen-transition metal complexes, including salen- manganese (III) complexes (hereafter abbreviated to salenMn(lll) complexes). US 6,177,419 describes various bipyridine manganese complexes. These compounds are disclosed as being superoxide dismutase and/or catalase and/or peroxidase mimetics and, accordingly, have antioxidant and/or free radical properties.

SalenMn(lll) complexes are known to give effective protection to cells, tissues and animals against oxidative damage ^{(2, 3}' ^{35, 36)}. The cationic complexes [(Λ/,Λ/'-bis(salicylidene)ethylenediamine)Mn(lll)]⁺ (hereafter compound 1) and [(/V,/V'-bis(3- methoxysalicylidene)ethylenediamine)Mn(lll)]⁺ (hereafter compound 2) have been shown to work as superoxide dismutase and catalase mimetics, protecting against free radical related disease in animals ⁽¹⁾. Compounds 1 and 2 are also known as EUK-8 and EUK-134 and their structures, as the chloride salts, are shown below.

These compounds are able to prevent formation of the hydroxyl radical through catalytic removal of the reactive oxygen species (ROS) superoxide and hydrogen peroxide. They therefore offer potential benefit to conditions ranging from normal ageing ⁽²⁾ to amyelotrophic lateral sclerosis ⁽³⁾, in which ROS are thought to be a contributing factor. The present inventors have been interested in the potential benefit of these compounds in treatment of ataxia-telangiectasia (A-T), a multifactorial autosomal recessive disease characterised by ionising radiation sensitivity ⁽⁴⁾, constitutive oxidative stress ^(5"8), and progressive cerebellar ataxia ^(9,10).

Although trials have shown the efficacy of compound 1 as a catalytic ROS scavenger in vivo (e.g. ⁽²⁾), it is also known that it cleaves plasmid DNA ^(11"13). Thus A-T cells, which are hypersensitive to double-strand DNA cleavage ⁽¹⁴⁾, might be unusually sensitive to any adverse effect of compound 1. A study of ischaemic response to compound 1 and its derivative 3,3'-methoxysalenMn(lll) (compound 2) strongly suggests that the latter is a more effective treatment for a condition where ROS are major contributors to the injury ⁽¹⁾.

While salenMn(lll) catalysts have useful antioxidant properties, protecting against free radical related diseases in animals, they can also be potent pro-oxidants which damage free DNA. Thus, despite the potential benefits, there are serious concerns over the use of such products as therapeutic agents. The present invention results from the discovery of a way of protecting against the pro-oxidant activity of salenMn(lll) complexes or derivatives, and more specifically the discovery of the synergistic effects of thiol (or reducible disulphide)/salen-manganese (III) compound coadministration, and which it is expected will make them safe enough for commercial use in drug therapy. The invention is also applicable to other salen-metal complexes.

According to a first embodiment the present invention provides a method for improving the safety and efficacy of catalytic antioxidant salen-metal complexes, such as salenMn(lll) complexes, by combining them with a compound that enhances intracellular thiol levels. Examples of such compounds include reducible disulphide or thiol compounds. For example, the salenMn(lll) complexes, such as compounds 1 and 2 above, may be combined with, for example, glutathione or α-lipoic acid. The latter, an antioxidant reducible disulphide dietary supplement, has the following structure 3:-

Salen-metal complexes suitable for use in the method of this invention include those having the formula:-

where M is Mn, Cu, Co, Fe, V or Cr; A is an anion;

R-i and R ' are independently H or alkyl groups; and

R₃, R₄ and R5 are independently H, alkyl, alkoxy, halide, OH, SH, NH₂ or

N0₂ groups. Preferably, M is Mn; at least one of R-i and R-i' is lower (Cι-C₆) alkyl, most preferably -CH₃ ; and at least one of R₃, R and R₅ is lower (C-i.Cβ) alkoxy, most preferably -OCH₃. The presence of a -CH₃ group at the R-i' position is particularly preferred since this has been found to increase the solubility of the product and, in consequence, makes it possible to use compounds (such as those containing the transition metals Cr, V, Fe, Co or Cu) that would otherwise have solubility properties that are unsuitable for use in pharmaceutical applications.

The salen-metal complexes may be combined with, for example, glutathione or α-lipoic acid either by adding the separate components simultaneously or in rapid succession to a preparation. It is to be understood that the salen-metal complexes may also be conjoined to the glutathione or α-lipoic acid.

According to a second embodiment of this invention there is provided a compound which comprises a catalytic antioxidant salen-metal complex combined with a compound that enhances intracellular thiol levels. The latter is typically a reducible disulphide or thiol compound, and most preferably, glutathione or α-lipoic acid. Examples of such hybrid structures are as follows:-

(counterion is an anion, typically chloride)

wherein M is Mn, Co, Fe, V, Cr or Cu; A is an anion; Ri, R-i' and R2 are independently H or alkyl groups; and

R_3> R₄ and R₅ are independently H, alkyl, alkoxy, halide, OH, SH, NH₂ or

NO₂ groups.

Preferably, M is Mn; A is chloride; at least one of R-i, R-i' and R₂ is lower (Ci-Cβ) alkyl, most preferably -CH₃; and at least one of R₃, R₄ and R₅ is lower (C-i-Cβ) alkoxy, most preferably -OCH₃.

According to a third embodiment of this invention there is provided a pharmaceutical composition which comprises a catalytic antioxidant salen- metal complex, such as a salenMn(lll) complex or derivative, together with a compound that enhances intracellular thiol levels. The latter compound is typically reducible disulphide or thiol compound, such as glutathione or α-lipoic acid. A drug therapy that safely and effectively reduces damage caused by reactive oxygen species, while avoiding the aforementioned dangers previously associated with the use of salenMn(lll) complexes, might reasonably be expected to extend human lifespan and also protect against a range of degenerative diseases, including diabetes, neurodegenerative disorders and anti-immune disease.

Since compound 1 prevents oxidative damage to cells, but causes damage to free DNA, this implies that common cellular constituents may interact with compound 1 to modify its activity. The tripeptide, reduced glutathione (GSH) is a major scavenging molecule present at millimolar levels within the cell. The present inventors have surprisingly and advantageously shown that GSH substantially curtails the DNA-damaging activity of compound 1. GSH is not efficiently taken up by cells, and a number of related molecules have been used to boost intracellular GSH levels. α-Lipoic acid (compound 3, also called thiotic acid or 2-dithiolane-3- pentanoic acid) is an essential cofactor of pyruvate dehydrogenase ⁽¹⁵⁾, which is taken up into cells by the same transporter as pantothenic acid ⁽¹⁶⁾. It is reduced within the cell (by NADH or lipoamide dehydrogenase) to break the 5-membered ring. Both the oxidised and reduced forms have antioxidant activity ^{(17 8)}. The reduced form can recycle scavengers such as ascorbate, can restore intracellular GSH levels ⁽¹⁹⁾ by recovery of cysteine from cystine, and can regenerate critical thiols. There is evidence that α-lipoic acid may ameliorate conditions including type I and type II diabetes ⁽²⁰' ²¹⁾, ischemia-reperfusion injury, neurodegenerative diseases ⁽⁵⁾, radiation injury and heavy metal poisoning ⁽¹⁸⁾.

ROS-related DNA damage can conveniently be measured by the comet assay (single cell gel electrophoresis) ^(22"25 Mammalian cells are embedded in an agar layer on a microscope slide, lysed, and subjected to alkaline electrophoresis. DNA containing strand breaks streams from the nucleus in a comet tail, whereas undamaged DNA remains trapped. The inventors have previously adapted this technique to study the pro-oxidant activity of GSH against free DNA ⁽²⁶⁾ by rinsing away the lysis mixture and incubating with GSH before electrophoresis. Although this technique can readily demonstrate DNA strand breakage, it does not indicate whether the pro-oxidant binds to the DNA before cleaving it. They have adapted the quartz crystal resonance sensor (OCRS) ⁽²⁷- ²⁸⁾ to address this question ⁽²⁹⁾. DNA is attached to the surface of a quartz crystal, which resonates at a characteristic frequency. A solution of the molecule of interest is passed over the crystal in a suitable flow chamber. If the molecule binds to the DNA, the resonance frequency decreases, if it cleaves DNA, the frequency is raised. Initially, the inventors investigated the differences between compounds 1 and 2 using this methodology ⁽²⁹⁾. It was observed that compound 1 bound tightly to the DNA, cleaved it, and finally intercalated in a weak fashion. This is in good agreement with previous reports in which DNA is incubated with compound 1 and an oxidant ⁽¹¹⁾, however, the present inventors' OCRS experiments used no such oxidant. Compound 2 only exhibited weak binding to the DNA throughout the experiment, without evidence of cleavage, which may explain why it is the more efficient superoxide dismutase/catalase mimetic of the two.

Compound 1 is a strong, and compound 2 a weak, pro-oxidant to free

DNA ^(13ι29). It therefore appears that these complexes interact with cellular constituents in a way that can block the pro-oxidant activity without preventing catalytic protection. The present inventors have shown that glutathione does indeed interact with salenMn(lll) complexes, preventing the binding of compound 1 to DNA, largely blocking its pro-oxidant activity.

As well as being an essential cofactor for pyruvate dehydrogenase, α-lipoic acid is an antioxidant in both its oxidised and reduced forms ⁽¹⁷⁾, which may also boost intracellular glutathione levels by releasing cysteine from cystine ⁽¹⁹⁾. Since it is actively transported into cells ⁽¹⁶⁾, it has the potential for use as a dietary supplement. It has been shown by the present inventors that the oxidised form of α-lipoic acid largely prevents binding of compound 1 to DNA and gives some protection against pro-oxidant activity, though less than glutathione. It also extends the catalytic activity of both compounds 1 and 2 against H₂0₂ in vitro. α-Lipoic acid does not offer cells significant protection against H₂0₂, but it does appear to enhance the protection offered by compound 2.

Catalytic antioxidants offer a number of potential advantages over conventional free radical scavengers. They should be far more effective, since one molecule can inactivate many potentially damaging reactive oxygen species. They offer supplementation of key protective enzymes, which cannot readily be achieved in other ways. Moreover, they inactivate H₂0 and superoxide, thus removing the precursors of the reactive species such as the hydroxyl radical against which scavengers protect. The salenMn(lll) group of compounds, however, have a clear ability to act as pro-oxidants, generating the reactive oxygen species against which protection is required. The present results indicate that a beneficial interaction with intracellular thiols may reduce this hazard within the cell or the animal, and that combined treatment with these molecules and thiol supplementation would be beneficial.

The present invention will now be further illustrated by the following example and with reference to the accompanying drawings, in which:

Figure 1 shows: A DNA strand breakage as indicated by comet length when human fibroblasts or free DNA are incubated with compounds 1 or 2. Left set: fibroblasts incubated with compounds 1 or 2; Centre set: fibroblasts incubated with H2O2 and compounds 1 or 2; Right set: free DNA incubated with compounds 1 or 2. B. Free DNA incubated with compounds 1 or 2, in the presence of GSH or α-lipoic acid.

Figure 2 shows the change in resonance frequency of a DNA-coated quartz crystal, following a solution of compound 1 being passed over it, in the presence or absence of GSH or α-lipoic acid.

Figure 3 shows the persistence of catalase-like activity of in the presence or absence of 50 μmol/L GSH or α-lipoic acid. (A) * 5 μmol/L 1 ; — ^D — ; 5 μmol/L 1 + 50 μmol/L α-lipoic acid; ° 5 μmol/L 1 + 50 μmol/L GSH; __ 5 μmol/L 2. (B) • 5 μmol/L 2 — π— 5 μmol/L 2 + 50 μmol/L α-lipoic acid; "' ' ° 5 μmol/L 2 + 50 μmol/L GSH; -—__ 5 μmol/L 1. Successive pulses of approximately 10 μmol/L H2O2 were added and the rate of removal of H2O2 determined after each pulse. Each treatment was repeated 9 times. Rate of removal is plotted against total H2O added. For clarity of presentation, values for adjacent H2O2 readings have been averaged and grouped.

Figure 4 shows the effects of compound 1 , compound 2 and α-lipoic acid on H2θ2-induced DNA strand breakage in human fibroblasts. Damage expressed as increase over control comet length

Figure 5 shows a comparison of DNA binding with different salenMn complexes.

EXAMPLE

Materials and Methods

Syntheses and reagents

Compounds 1 and 2 were synthesised by reaction of ethylenediamine with salicylaldehyde or o-vanillin respectively ⁽²⁹'³°⁾. Culture media were obtained from Life Sciences (Paisley, UK). Standard chemicals were from Sigma Aldrich (Poole, Dorset, UK).

Preparation of Hybrid Molecules (Compounds 4 and 5)

Compound 4 may be prepared in situ through combination of an aqueous solution of the chloride salt of the parent salen manganese compound and

α-lipoic acid. Alternatively, derivatives may be prepared by addition of the

parent salen manganese compound to the sodium salt of α-lipoic acid in

aqueous solution to form a solid precipitate.

An example of compound 5, in which Ri is H and R is -CH₃, may be prepared in the following manner. Ethylenediamine was added to a methanolic solution of 2-hydroxy-5-methyl-1 ,3-benzenedicarboxaldehyde, in the ratio 1 :2, and stirred for 2 hours. A solution of manganese (II) acetate was added with stirring to give a brown solution and stirring continued for 24 hours. Following methanol evaporation, a brown precipitate of the manganese (II) complex formed and was filtered off. The manganese (II) complex was dissolved in hot water, stirred and a saturated potassium chloride solution added dropwise over 10 minutes to oxidise the complex. The solution cooled to room temperature, the precipitate was filtered off and washed with a small amount of water to give the manganese (III) complex as the chloride. The product was dissolved in methanol and a methanolic solution of 6-thiotic amine^* added. The mixture was stirred for a further 24 hours. The precipitate formed was filtered and washed with methanol to give compound 5 for which Ri is H and R₂ is - CH₃.

^*6-Thiotic amine was prepared through the addition of sodium hydroxide and 10% sodium hypochloride to a solution of 6-thiotic amide. The solution was heated to 80°C for 2 hours, allowed to cool to 50°C and 10% aqueous sodium bisulfite added resulting in the precipitation of 6-thiotic amine as a yellow solid.

Cell culture and comet assay

Primary human fibroblasts (1 BR.3 ⁽³¹⁾) were cultured in Eagle's minimal essential medium with 15% fetal calf serum as previously described ⁽³²⁾. Comet assays were by the procedure of Singh ⁽²²⁾, with modifications. Briefly, using a second slide, a series of standard clear microscope slides were spread with 45 μl 0.3% agarose in H₂0 and allowed to dry overnight ⁽³³⁾. Fibroblast cultures were trypsinised, 2 x 10⁴ cells suspended in 50 μl of 0.7% NuSeive low melting point agarose were placed on the slide, and a coverslip added. The coverslip was removed, appropriate volumes of the test substances were placed on the surface of the agar, the coverslip was replaced, and the slide incubated at 37°C over damp tissue (to prevent drying). After 1 hour, the coverslip was removed and the slide was placed in lysis mixture (2.5 mol/L NaCI, 200 mmol/L NaOH, 100 mol/L EDTA-Na₂, 10 mmol/L TRIS base, 10% v/v dimethylsulfoxide, 1 % v/v Triton X-100) at 4^°C for at least 1 h. Slides were transferred to electrophoresis buffer (300 mmol/L NaOH, 1 mmol/L EDTA-Na2), incubated at 15°C for 40 min and 20 V were applied for 24 min. To treat free DNA, the lysis mixture was rinsed away with Dulbecco's phosphate buffered saline (PBS), and the mixture under test added in PBS ⁽²⁶⁾. After incubation (60 min at 37°C) the slides were placed in electrophoresis buffer and subjected to immediate electrophoresis (20 V, 24 min) without the unwinding step.

QCRS

The quartz crystal resonance sensor was used to measure DNA interactions as described previously ⁽²⁹⁾. Briefly, 10 MHz quartz crystals (Hi-Q International, Cambridge, UK) were coated with 100 nm thickness gold electrodes, and placed in a special flow chamber (20 μl volume), attached to a gain control oscillator. Resonance frequency was determined once per second with a Fluka-6689 resonance counter. Via the flow chamber, the crystals were washed with ethanol, coated with neutravidin, and then biotinylated DNA. After each treatment, the chamber was flushed with Sorenson's phosphate buffer until the resonance frequency was stable. A pulse of the test compound or mixture in Sorenson's buffer was then added, and the change in resonance frequency recorded.

H∑θ2 assay

A solution (5 μmol/L) of the test compound was placed in a spectrophotometric cell. H₂0₂ was added to give 10 μmol/L and the rate of loss of absorbance at 233 nm determined. Successive pulses of H₂0 were added, and the rate of removal measured each time.

Results

Human fibroblasts, when incubated for 1 hour with compounds 1 or 2, showed no evidence of DNA damage as determined by the comet assay, and both compounds substantially reduced the DNA damage caused by incubation with 5 μmol/L H₂0₂ for 20 min at 4^°C (Fig 1 A). However, when incubated with free DNA in the comet assay, compound 1 at 10 μmol/L caused extensive strand breakage (Fig 1A). Compound 2 gave much less damage, confirming the observations of Gravert and Griffin ⁽³⁴⁾. It was hypothesised that a common scavenging molecule might be responsible for the lack of damage when intact cells were treated with compound 1. The inventors therefore treated free DNA with compounds 1 or 2 in the presence of 1 mmol/L GSH. In each case, the amount of strand breakage was substantially reduced (Fig 1 B) and was lower than the damage seen with either compound alone. (GSH is a weak pro-oxidant in this assay ⁽²⁶⁾). GSH is not taken up efficiently by cells, and it is common to use other molecules to achieve indirect supplemention of intracellular GSH. The inventors therefore tested α-lipoic acid, as a molecule which can raise intracellular GSH levels. This also gave protection against the pro-oxidant activity of compounds 1 and 2, but was less effective than GSH. Within the cell it would be present in the more effective reduced form ^{17). GSH also lowered the amount of damage seen when free DNA was incubated with compound 2 (Fig 1 B).

Although the comet assay can show strand breakage, it does not give an indication of whether DNA binding is involved in the damage process. The inventors had previously used QCR to show (29) that compound 1 , but not compound 2 binds strongly to DNA, before cleaving it. They therefore passed compound 1 over a DNA-coated quartz crystal in the presence of GSH or α-lipoic acid . Fig 2 shows that both compounds reduced binding of compound 1 to DNA on a quartz crystal, and prevented strand breakage from occurring. The inventors next tested the effect of GSH and α-lipoic acid on the catalytic activity of compounds 1 and 2. Under the assay conditions, H₂O₂ is rapidly removed on incubation in the presence of compound 1 , however, when successive pulses of H₂O2 are added to compound 1 , the rate of removal rapidly declines (Fig 3A). Compound 2 is initially a less effective catalyst than compound 1 , in contrast to the report of Baker et al. ⁽¹⁾, but its activity shows greater persistence (Figs 3A and 4B). When a x10-fold excess of α-lipoic acid is added together with compound 1 , the initial rate of removal is lower, and the overall persistence of activity is broadly similar to that of compound 2. GSH shows little effect on the removal of H₂0₂ by compound 1. In contrast, α-lipoic acid has little effect on the activity of compound 2, whereas GSH increases the initial activity of 2 while giving lower activity at later times (Fig 3B). Thus, in the presence of GSH, compound 2 behaves similarly to compound 1.

Finally, the inventors tested whether α-lipoic acid would influence the protection of cells against H₂O₂ by compounds 1 or 2. From Fig 4 it can again be seen that neither of the agents induced damage in the absence of H₂0₂. α-Lipoic acid itself did not protect against H₂0₂ but compounds 1 and 2 both gave partial protection, somewhat less than in the earlier series of experiments (see Fig 3A). α-Lipoic acid slightly enhanced the protection given by compound 2, but not compound 1.

A specific example of compound 5 is compound 7 in which Ri is H, Ri' is -CH₃ and R₂ is -CH₃. Compound 7 has been prepared as described, and has been tested by the inventors for catalytic activity against H₂O₂. Preliminary data indicate that it is approximately twice as active as the combination of compound 2 and α-lipoic acid. Electrochemical tests, by analogy with literature methods ⁽³⁷⁾, show enhanced antioxidant activity over compounds 1 and 2.

A further example of compound 4 in which Ri is H and R-i' is CH₃ and R_3, R₄ and R₅ are each H has been prepared and tested using the OCRS and comet assay techniques. This compound exists in two chiral forms (designated compounds 6a and 6b). Figure 5 shows the DNA binding and cleaving capacity of compound 6b, compared to that of compounds 1, 2, 6a and 7. Both compounds 6a and 6b show diminished binding to DNA in comparison with compound 1 (Figure 5). Compound 6a still retains a limited capacity to cause strand breakage. In contrast to compounds 1 and 2, which show DNA binding and cleavage, and binding only, respectively, both DNA binding and cleavage capacity are effectively absent in compound 6b as judged by the OCRS technique.

Introduction of the CH₃ group, in particular at the R-i' position, additionally increases the aqueous solubility of these compounds over their unsubstituted analogues. A further application of this substitution is to increase the solubility of candidate structures which may now incorporate metals that hitherto could not be used due to unfavourable solubility properties. Examples would be compounds 4 and 5 where the transition metal was typically Cr, V, Fe, Co or Cu instead of Mn. References

1. Baker, K., Marcus, C. B., Huffman, K., Kruk, H., Malfroy, B., and

Doctrow, S. R. (1998) Journal of Pharmacology and Experimental

Therapeutics. 284(1), 215-221 2. Melov, S., Ravenscroft, J., Malik, S., Gill, M. S., Walker, D. W.₍

Clayton, P. E., Wallace, D. C, Malfroy, B., Doctrow, S. R., and

Lithgow, G. J. (2000) Science 289, 1567-1569 3. Jung, C. W., Rong, Y. Q., Doctrow, S., Baudry, M., Malfroy, B., and

Xu, Z. S. (2001) Neuroscience Letters 304(3), 157-160 4. Taylor, A. M. R., Harnden, D. G., Arlett, C. F., Harcourt, S. A.,

Lehmann, A. R., Stevens, S., and Bridges, B. A. (1975) Nature 258,

427-429 5. Gatei, M., Shkedy, D., Khanna, K. K., Uziel, T., Shiloh, Y., Pandita, T.

K., Lavin, M. F., and Rotman, G. (2001) Oncogene 20(3), 289-294 6. Barlow, C, Dennery, P. A., Shigenaga, M. K., Smith, M. A., Morrow, J.

D., Roberts, L. J. n., Wynshaw-Boris, A., and Levine, R. L. (1999)

Proceedings of the National Academy of Sciences of the United

States of America 96(17), 9915-9919

7. Kamsler, A., Daily, D., Hochman, A., Stem, N., Shiloh, Y., Rotman, G., and Barzilai, A. (2001 ) Cancer Research 61 (5), 1849-1854

8. Quick, K. L., and Dugan, L. L. (2001 ) Annals of Neurology 49(5), 627- 635

9. Vintners, H. V., Gatti, R. A., and Rakic, C. (1985) in Ataxia- Telangiectasia: Genetics, Neuropathology and Immunology of a Degenerative Disease of Childhood (Gatti, R. A., and Swift, M., eds), pp. 233-255, Alan R. Liss, New York 10. Gatti, R. A., and Vintners, H. V. (1985) in Ataxia-Telangiectasia: Genetics, Neuropathology and Immunology of a Degenerative Disease of Childhood (Gatti, R. A., and Swift, M., eds), pp. 225-232, Alan R. Liss, New York 11. Gravert, D. J., and Griffin, J. H. (1993) Journal of Organic Chemistry 58, 820-822

12. Griffin, J. H. (1995) Bioorganic & Medicinal Chemistry Letters 5(1 ), 73- 76

13. Gravert, D. J., and Griffin, J. H. (1996) in Metal Ions in Biological Systems Vol. 33, pp. 515-536

14. Foray, N., Priestley, A., Alsbeih, G., Badie, C, Capulas, E. P., Arlett, C. F., and Malaise, E. P. (1997) International Journal of Radiation Biology 72, 271-283

15. Lehninger, A. L., Nelson, D. L., and Cox, M. M. (1993) Principles of Biochemistry, 2nd Ed., Worth Publishers, New York

16. Prasad, P. D., and Ganapathy, V. (2000) Current Opinion in Clinical Nutrition and Metabolic Care 3(4), 263-266

17. Packer, L, Witt, E. H., and Tritschler, J. (1995) Free Radical Biology and Medicine 19(2), 227-250 18. Packer, L., Tritschler, H. J., and Wessel, K. (1997) Free Radical

Biology and Medicine 22(1/2), 359-378 19. Han, D., Handelman, G., Marcocci, L., Sen, C. K., Roy, S., Kobuchi,

H., Tritschler, H. J., Flohe, L., and Packer, L. (1997) Biot actors 6(3),

321-338 20. Maddux, B. A., See, W., Lawrence, J. C, Goldfine, A. L., Goldfine, I.

D., and Evans, J. L. (2001) Diabetes δO, 404-410 21. Konrad, D., Somwar, R., Sweeney, G., Yaworsky, K., Hayashi, M., Ramlal, T., and Klip, A. (2001) Diabetes 50, 1464-1471

22. Singh, N. P., McCoy, M. T., Tice, R. R., and Schneider, E. L. (1988) Experimental Cell Research 175, 184-191 23. Collins, A. R., Duthie, S. J., and Dobson, V. L. (1993) Carcinogenesis

14, 1733-1735 24. Collins, A., Dusinska, M., Franklin, M., Somorovska, M., Petrovska,

H., Duthie, S., Fillion, L., Panayiotidis, M., Raslova, K., and Vaughan,

N. (1997) Environmental and Molecular Mutagenesis 30(2), 139-146 25. Clingen, P. H., Lowe, J. E., and Green, M. H. L. (2000) in Aging:

Methods and Protocols (Barnett, Y. A., and Barnett, C. R., eds) Vol.

38, pp. 143-157, Humana Press Inc., Totowa, NJ

26. Thomas, S., Lowe, J. E., Hadjivassiliou, V., Knowles, R. G., Green, I. C, and Green, M. H. L. (1998) Biochemical and Biophysical Research Communications 243(1 ), 241 -245

27. Pavey, K. D., Olliff, C. J., and Paul, F. (2001) Analyst Communications 126, 426-428

28. Pavey, K. D., AN, Z., Olliff, C. J., and Paul, F. (1999) Journal of Pharmaceutical and Biomedical Analysis 20(1 -2), 241 -245 29. Fucassi, F., Pavey, K. D., Lowe, J. E., Olliff, C. J., Green, M. H. L.,

Paul, F., and Cragg, P. J. (2001) Chemical Communications , 841-842

30. Pecoraro, V. L, and Butler, W. M. (1986) Acta Cryst. C42, 1151 -1154

31. Arlett, C. F., and Harcourt, S. A. (1980) Cancer Research 40, 926-932

32. Woollons, A., Clingen, P. H., Price, M. L., Arlett, C. F., and Green, M. H. L. (1997) British Journal of Dermatology 137, 687-692 33. Klaude, M., Eriksson, S., Nygren, J., and Ahnstrom, G. (1996) Mutation Research 363(2), 89-96

34. Gravert, D. J., and Griffin, J. H. (1996) Inorganic Chemistry 35(17), 4837-4847 35. Bruce,A.J.,alfroy, B., and Baudry, M. (1996) Proceedings of the

National Academy of Sciences of the United States of America 93(6),

2312-2316 36. Henke, S. L. (1999) Expert Opinion on Therapeutic Patents 9(2), 169-

180 37. The voltammetric behaviour of superoxide dismutase/catalase mimics,

D. O'Hare, F. Fucassi, P. J. Cragg and M H L Green, Electroanalysis, in press.

Claims

1. A method for improving the safety and efficacy of catalytic antioxidant salen-metal complexes which comprises combining them with a compound that enhances intracellular thiol levels.

2. A method as claimed in claim 1 , wherein the salen-metal complex has the formula:-

A

where M is Mn, Cu, Co, Fe, V or Cr; A is an anion;

R-i and Ri' are independently H or alkyl groups; and R₃, R and R₅ are independently H, alkyl, alkoxy, halide, OH, SH, NH₂ or N0₂ groups.

3. A method as claimed in claim 2, wherein M is Mn, A is chloride, at least one of Ri, Ri' and R is -CH₃ and at least one of R₃, R₄ and R₅ is -OCH₃.

4. A method as claimed in any one of claims 1 to 3, wherein the compound that enhances intracellular thiol levels is glutathione or α-lipoic acid.

5. A compound which comprises a catalytic antioxidant salen-metal complex combined with a compound that enhances intracellular thiol levels.

6. A compound as claimed in claim 5, wherein the compound that enhances intracellular thiol levels is a reducible disulphide or thiol compound.

7. A compound as claimed in claim 5 or claim 6, wherein the compound that enhances intracellular thiol levels is glutathione or α-lipoic acid.

8. A compound as claimed in any one of claims 5 to 7, which has the formula:-

wherein M is Mn, Co, Fe, V, Cr or Cu;

A is an anion;

R-i, R-i' and R₂ are independently H or alkyl groups.

9. A pharmaceutical composition which comprises a catalytic antioxidant salen-metal complex or derivative thereof and a compound that enhances intracellular thiol levels.

10. A pharmaceutical composition as claimed in claim 9, wherein the salen-metal complex comprises a salen Mn(lll) complex or derivative and the compound that enhances intracellular thiol levels comprises glutathione or α-lipoic acid.