Glutahone in Parkmson’s Disease: A Lnk Between &dative Stress and Mitochondrial Damage? Donato A. Di Monte, MD, Piu Chan, MD, PhD, and Martha S. Sandy, PhD
Several links exist between the two mechanisms of neuronal degeneration (i.e., oxygen radical production and mitochondrial damage) proposed to have a role in Parkinson’s disease. Indeed, mitochondria are critical targets for the toxic injury induced by oxygen radicals, and experimental evidence suggests that mitochondrial damage may cause an increased generation of oxygen radicals. A potentially important link between these two mechanisms of neurodegeneration is glutathione. Because of the scavenging activity of glutathione against accumulation of oxygen radicals, its decrease in the brains of parkinsonian patients has been interpreted as a sign of oxidative stress; however, this change may also result from or lead to mitochondrial damage. It is conceivable therefore that regardless of whether oxidative stress or mitochondrial damage represents the initial insult, these toxic mechanisms may both contribute to neuronal degeneration via changes in glutathione levels. Di Monte DA, Chan P, Sandy MS. Glutathione in Parkinson’s disease: a link between oxidative stress and mitochondrial damage? Ann Neurol 1992;32:Slll-S115
Two main hypotheses are currently debated on the mechanism of neuronal degeneration underlying Parkinson’s disease. The first hypothesis involves formation of highly reactive oxidizing species (i.e., oxygen radicals), which may lead to neuronal damage by shifting the cellular oxidation-reduction (redox) equilibrium toward oxidation (so-called oxidative stress) 11). The second and more recent hypothesis on the mechanism of neurodegeneration in Parkinson’s disease concerns deficiency of mitochondrial activity and the consequent failure of energy production and changes in neuronal metabolism 12). In this article, we emphasize possible relationships between these two pathogenetic hypotheses. In particular, changes in brain levels of glutathione are discussed in terms of their significance as markers of oxidative stress and/or mitochondrial damage in Parkinson’s disease. Mechanisms of Neuronal Degeneration in Parkinson’s Disease Elucidating the time course of toxic events during cell degeneration is a critical but elusive process. Indeed, relatively early or late occurrence of a metabolic or toxic change may provide a clue regarding its role in the development of tissue injury and thus in the pathogenesis of a disease. When dealing with Parkinson’s disease, the need to establish a temporal relationship between pathological events may bear somewhat dis-
From the California Parkinson’s Foundation, San Jose, CA.
couraging implications. How can we expect to identify “early” metabolic or toxic changes if we assume (as is currently believed 13)) that the neurodegeneration underlying Parkinson’s disease has already reached an advanced stage at initial diagnosis? It may well be that most of the changes observed in tissues of parkinsonian patients are the results rather than the actual causes of neurodegeneration. The initial noxa (or the initial noxae) may or may not still be active at diagnosis and might have occurred long before clinical expression of the disease. A significant step toward better understanding of the temporal and causal relationship of toxic events in Parkinson’s disease would be identification of early markers of disease development, and it is promising that major efforts are being focused on achieving this goal by a number of research groups 14). The previous concepts and concerns need to be taken into consideration when evidence supporting either the oxidative stress or the mitochondrial damage hypotheses is discussed. For example, increased levels of lipid peroxidation products in the substantia nigra of parkinsonian patients C5-j may be interpreted as a sign of the role of oxidative stress in neurodegeneration, but they could merely be the result of peroxidative processes following neuronal damage 161. Likewise, the reported decrease in the activity of enzymes of the respiratory chain in muscle tissue of parkinsonian patients 17, 81 may or may not be a marker of genetic predisposition to degenerative processes
Address correspondence to Dr Di Monte, California Parkinson’s Foundation, 2444 Moorpark Ave, Suite 3 16, San Jose, CA 95 128.
triggered by mitochondrial damage. Thus, caution is needed when occurrence of a metabolic or toxic change is interpreted in terms of its relevance as a pathogenic factor. This caution is most important if changes are found in postmortem specimens or in tissues not directly involved in the pathophysiology of Parkinson’s disease. Another reason for a cautious approach when debating the role of oxidative stress versus mitochondrial damage in Parkinson’s disease is emphasized by the following consideration. Although neurodegeneration may initially be triggered either by the formation of oxygen radicals or by an impairment of energy supplies by mitochondria, both of these toxic processes may be involved in the subsequent development of tissue injury because a relationship exists between oxidative stress and mitochondrial damage. Generation and accumulation of oxidizing species is likely to affect mitochondrial function and energy production. For example, exposure of cells to H202causes a depletion of adenosine triphosphate (ATP) 1111, and Ca” sequestration by mitochondria is impaired as a consequence of oxidative stress (via changes in the redox levels of glutathione and pyridine nucleotides) 19, 101. Primary mitochondrial damage can also result in increased generation of oxygen radicals, as demonstrated by studies in which addition of antimycin A or C N - (two blockers of mitochondrial electron flow) to mitochondria and submitochondrial particles caused oxygen radical formation and, consequently, DNA alterations C121. Thus, oxidative stress and mitochondrial damage may act together in determining neurodegeneration rather than being two conflicting hypotheses on the pathogenesis of Parkinson’s disease. The possible relationship between these two mechanisms of cell injury is emphasized by their association with changes in glutathione homeostasis. Glutathione and Parkinson’s Disease A key reaction against the accumulation of oxygen radicals, both under physiological conditions and as the result of a toxic insult, is the reduction of H202to H 2 0 , catalyzed by the enzyme glutathione peroxidase. Reduced glutathione (GSH) represents the electron donor in this reaction and it is oxidized to GSSG (oxidized glutathione). Unless overwhelming production of H202occurs, GSSG is rapidly reduced back to GSH by glutathione reductase at the expense of NADPH, and thus the tissue ratio of GSH to GSSG remains quite constant. One possible source of H202is related to the turnover of dopamine by monoamine oxidase (MAO), and accumulation of H202via this pathway has been suggested to be involved in degeneration of dopaminergic nigrostriatal neurons 111. Experimental evidence in support of a relationship between increased dopamine S112 Annals of Neurology
turnover, H202production, and oxidation of GSH to GSSG has been provided by the elegant work of Cohen and co-workers 1131. In one study from this research group, increased presynaptic turnover of dopamine was induced in mice by administration of reserpine, leading to an increase in GSSG levels in the striatum. If H202formation via M A 0 or if oxidative stress in general is involved in neurodegeneration in Parkinson’s disease, changes in brain glutathione may be expected to occur. Indeed, a decrease in total (GSH + GSSG) glutathione as well as in GSH levels has been reported in the brains of parkinsonian patients as compared to control subjects 114-161, and, in two of these studies 114, 151, this decrease appeared to be selective for the substantia nigra. These changes have been interpreted as the result of increased oxygen radical production in the nigrostriatal pathway of parkinsonian patients, and have led to the speculation that Parkinson’s disease may be attributable to nigral glutathione deficiency 1141. However, a number of important considerations related to measurements of glutathione in human tissues may suggest a more cautious interpretation of these results. First, although GSSG measurements represent a more reliable index of oxidative stress than total glutathione, changes in tissue levels of GSSG are likely to be only transient and may be difficult to interpret. As mentioned, GSSG is normally reduced back to GSH, and, even under conditions of overwhelming production of H202,the acute increase in GSSG levels would rapidly be reversed by excretion from cells and clearance from tissue. Furthermore, increased GSSG levels in autopsy specimens may be the result of postmortem oxidative processes. For example, in one postmortem study 1141, GSSG levels in all areas of parkinsonian as well as control brains were greater than GSH levels. This finding obviously has no physiological basis and probably represents an artifact of the long lag time between death, dissection of the brain, and processing of the samples. Postmortem changes can dramatically affect measurements of GSH as well. In a relatively simple experiment, we killed C57BL/6 mice (n = 5) by cervical dislocation and rapidly dissected the striatum from both sides of the brain. The striatum from one side was immediately immersed in perchloric acid and sonicated, and the acid-extracted GSH and GSSG were derivatized and measured by high-performance liquid chromatography 1171; in contrast, the striatum from the other side of the brain was left on ice for 1 hour before being processed as described. Results of this experiment reveal that striatal levels of GSH decreased rapidly as a consequence of postmortem processes; they were 35% lower when analysis was delayed by only 1 hour (Fig 1). The loss of GSH under these
Supplement to Volume 32, 1992
I GSH GSSG
Fig 1 . Postmortem changes in striatal levek of glutathione. Reduced glutathione (GSH) (open bars j and oxidized glutathione (GSSG) (striped barsj levels were measured either immediately (time 0) in the striatum from one side of the brain or 1 hour afer dissection of the striatum from the other side of the brain ofC57BLi6 mice (n = 5). Values are mean k SEM. Asterisk = statistically different (p < 0.05)from the corresponding value at time 0.
experimental conditions could not be accounted for by a stoichiometrical increase in GSSG, indicating that peptidase activity could also have a role in postmortem changes of glutathione levels in the brain. It is obvious that technical artifacts and postmortem changes represent major concerns when measurements of glutathione are performed on brain specimens obtained at various times after death (e.g., within 8 hours 1151 or 4 to 24 hours [l6]). Therefore, to lower the risk of erroneous interpretations of results, techniques of sample preparation and analysis should be as consistent as possible for both control and disease specimens. As mentioned, the decrease in glutathione levels seen in brain tissue of parkinsonian patients has been interpreted as a sign of oxygen radical production. Does this glutathione loss render the mitochondrial injury hypothesis less tenable? Is there any relationship between glutathione levels and mitochondrial activity? Glutathione, Oxidative Stress, and Mitochondrial Damage Although most glutathione is localized in the cytosolic fraction, approximately 10% of the total cellular gluta-
thione is compartmentalized within mitochondria [ 181. Because mitochondria also contain glutathione peroxidase, glutathione reductase, and NADPH (generated from NADH via transhydrogenation), a complete system for detoxifying hydroperoxides is present within these organelles. The mitochondrial pool of GSH is also likely to be involved in maintaining intramitochondrial protein thiols in the reduced state. These protein thiols are essential for a number of functions of these organelles, including selective membrane permeability and Ca” homeostasis. Thus, excessive production of H202within mitochondria may lead to depletion of mitochondrial GSH, oxidation of protein thiols, and impairment of mitochondrial function, providing a clear example of the relationship between glutathione status, oxidative stress, and mitochondrial damage. This relationship may have relevant implications in terms of the degeneration of dopaminergic neurons, because substrates of M A 0 may be sources of H,O, within mitochondria and may cause a decrease in mitochondrial GSH . Mitochondrial G S H originates from the cytosol [20}. Therefore, any significant decrease in cellular levels of glutathione is likely to affect mitochondrial function via loss of mitochondrial GSH. This toxic pathway is illustrated in Figure 2 and has recently been demonstrated experimentally by Jain and colleagues 1213. In this study, brain levels of glutathione were markedly decreased by administration of buthionine sulfoximine (BSO; an inhibitor of GSH synthesis) to newborn rats. This treatment caused mitochondrial damage as assessed by both electron microscopy and measurements of the activity of citrate synthase, a mitochondrial matrix marker enzyme. Mitochondria were enlarged and their structure was destroyed, and citrate synthase activity was significantly decreased. It is important to underline that BSO is not known to induce oxygen radical production; therefore, oxidative stress is not necessary to trigger loss of cytosolic GSH that leads to mitochondrial damage. Figure 2 suggests a novel interpretation of the results of glutathione measurements in parkinsonian brain, because lower levels of GSH in the substantia nigra may predispose to mitochondrial damage. If loss of GSH may cause mitochondrial damage, it is also conceivable that impairment of mitochondrial function can lead to a decrease in cytosolic GSH. GSH synthesis requires ATP, and thus a deficiency of energy supplies by mitochondria is likely to affect the cellular turnover of glutathione (Fig 3). Experimental support for this chain of events is provided by an in vitro study in which GSH and GSSG were measured in cells exposed to three mitochondrial poisons (i.e., potassium cyanide [KCN}, antimycin A, and l-methyl-4phenylpyridinium (MPP+) [22}. All compounds caused a decrease in GSH that was not accompanied Di Monte
GSH in PD S113
cytosolic glutathione depletion
mitochondrial GSH deficiency
mitochondrial GSH deficiency
impaired GSH synthesis
Fig 2. Relationship between reduced glutathione (GSHI depletion and mitochondrial damage.
mitochondria1 darnage mirochondrial damage
Fig 4. Cyclic toxic events linking oxygen radical production and mitochondrialdamage via changes in glutathione levels. ATP = adenosine triphosphate; GSH = reduced glutathione.
cytosolic glutathione depletion
Fig 3. Relationship between mitochondrial damage and reduced glutathione (GSH) depletion. ATP = adenosine triphosphate.
by a stoichiometrical increase in GSSG and could be counteracted by addition of substrates for glycolytic production of ATP. Thus, GSH loss induced by KCN, antimycin A, or MPP’ was not due to oxidation to GSSG and appeared to be correlated to intracellular levels of ATP. MPP+ is a metabolite and the ultimate mediator of l-methyl-4-phenyl-l,2,3,6-tetrahydropyridine (MPTP) neurotoxicity 123, 241. MPTP causes a parkinsonian syndrome in humans 1251 and nonhuman primates 1261,and its mechanism of action has been proposed as a model for metabolic and toxic events that might have a role in idiopathic Parkinson’s disease. Therefore, the mechanism by which MPP’ causes a decrease in GSH levels may be particularly relevant in interpreting the loss of glutathione seen in parkinsonian brain, and may point to deficiency of mitochondrial activity as a possible initial factor. Consequent impairment of glutathione turnover may then contribute to make dopaminergic neurons of the substantia nigra particularly susceptible to oxidative damage.
Conclusion It is clear that a number of links exist between oxidative stress and mitochondrial damage and that glutathione may represent one such link. If we combine the two toxic pathways in Figures 2 and 3, the resulting S114 Annals of Neurology
cycle (Fig 4) suggests that, regardless of the initial insult, a cascade of events involving both oxygen radicals and mitochondrial metabolism is likely to contribute to cell injury. In this respect, a rigid distinction between the oxidative stress and the mitochondrial hypotheses of neurodegeneration in Parkinson’s disease may only be artificial. It is also evident that interpretation of metabolic and toxic changes in parkinsonian brain requires careful evaluation. The time course of events, the occurrence of postmortem processes, and the interactions between toxic pathways are only some of the factors that need to be considered. When discussing glutathione changes in the brains of parkinsonian patients, we might also want to elucidate, for example, where these changes occur and which types of cells they might involve. Experimental data suggest that most glutathione in the brain is localized in glial cells rather than within neurons 127, 281, raising the intriguing possibilities that (1) loss of tissue glutathione may reflect changes in the proportion of glial versus neuronal cells, or (2) glial cells may have a more active role in human neurological disorders than previously suspected. The possible involvement of glial cells in neurodegenerative processes is only one example of relevant implications that may stem from the available data and have yet to be investigated in detail. The ability to recognize all such implications is essential to direct successfully future efforts in this rapidly developing research field. This work was supported by the California Parkinson’s Foundation. The authors wish to thank Dr Sarah A. Jewel1 for her comments on the manuscript.
Dr Chan is a recipient of the Bonomley Research Fellowship from the California Parkinson’s Foundation.
Supplement to Volume 32, 1992
References 1. Olanow CW. Oxidation reactions in Parkinson’s disease. Neurology 1990;40(suppl 3):32-37 2. Di Monte DA. Mitochondrial DNA and Parkinson’s disease. Neurology 1991;41(suppl 2):38-42 3. Langston JW. Predicting Parkinson’s disease. Neurology 1990;40(~~ppl 3):70-74 4. Langston JW, Koller WC. The next frontier in Parkinson’s disease: presymptomatic detection. Neurology 1991;41(suppI 2): 5-7 5. Dexter DT, Carter CJ, Wells FR,et al. Basal lipid peroxidation in substantia nigra is increased in Parkinson’s disease. J Neurochem 1989;52:381-389 6. Smith MT, Sandy MS, Di Monte D. Free radicals, lipid peroxidation, and Parkinson’s disease. Lancet 1987;1:38 7. Shoffner JM, Watts RL,Juncos JL, et al. Mitochondrial oxidative phosphorylation defects in Parkinson’s disease. Ann Neurol 1991;30:332-339 8. Bindoff LA, Birch-Machin M, Cartlidge NEF, et al. Mitochondrial function in Parkinson’s disease. Lancet 1989;2:49 9. Bellomo G , Jewel1 SA, Thor H, Orrenius S. Regulation of intracellular calcium compartmentation: studies with isolated hepatocytes and t-butyl hydroperoxide. Proc Natl Acad Sci USA 1982;79:6842-6846 10. Di Monte D, Bellomo G, Thor H, et al. Menadione-induced cytotoxicity is associated with protein thiol oxidation and alteration in intracellular Ca2+homeostasis. Arch Biochem Biophys 1984;235:343-350 11. Schraufstatter IU, Hyslop PA, Hinshaw DB, et al. Hydrogen peroxide-induced injury of cells and its prevention by inhibitors of poly(ADP-ribose) polymerase. Proc Natl Acad Sci USA 1986;83:4908-4912 12. Bandy B, Davison AJ. Mitochondrial mutations may increase oxidative stress: implications for carcinogenesis and aging? Free Radic Biol Med 1990;8:523-539 13. Spina MB, Cohen G. Dopamine turnover and glutathione oxidation: implication for Parkinson’s disease. Proc Natl Acad Sci USA 1989;86:1398-1400 14. Perry TL, Godin DV, Hansen S. Parkinson’s disease: a disorder due to nigral glutathione deficiency? Neurosci Lett 1982;33: 305-310 15. Perry TI.,Yong VW. Idiopathic Parkinson’s disease, progressive supranuclear palsy and glutathione metabolism in the substantia nigra of patients. Neurosci Lett 1986;67:269-274
16. Riederer P, Sofic E, Rausch W, et al. Transition metals, ferritin, glutathione, and ascorbic acid in parkinsonian brain. J Neurochem 1989;52:515-520 17. Reed DJ, Babson JR, Beatty PW, et d.High-performance liquid chromatography analysis of nanomole levels of glutathione, glutathione disulfide, and related thiols and disulfides. Anal Biochem 1980;16055-62 18. Reed DJ. Glutathione: toxicological implications. Annu Rev Pharmacol Toxic01 1990;30:603-63 1 19. Sandri G, Panfili E,Ernster L. Hydrogen peroxide production by monoamine oxidase in isolated rat-brain mitochondria: its effect on glutathione levels and Ca2‘ efflux. Biochem Biophys Acra 1 9 9 01035:300-3 05 20. Mktensson J, Lai JCK, Meister A. High-affinity transport of glutathione is part of a multicomponent system essential for mitochondrial function. Proc Natl Acad Sci USA 1990;87: 7185-7189 21. Jain A, Mktensson J, Stole E, et al. Glutathione deficiency leads to mitochondrial damage in brain. Proc Natl Acad Sci USA 1991;88:1913-1917 22. Mithofer K, Sandy MS, Smith MT, Di Monte D. Mitochondrial poisons cause depletion of reduced glutathione in isolated hepatocytes. Arch Biochem Biophys 1992;295: 132136 23. Nicklas WJ, Vyas I, Heikkila RE. Inhibition of NADH-linked oxidation in brain mitochondria by l-methyl-4-phenylpyridine, a metabolite of the neurotoxic l-methyl-4-phenyl-l.2,3,6-tetrahydropyridine. Life Sci 1985;36:2503-2508 24. Bradhury AJ, Costall B, Domeney AM, et al. l-Methyl-4phenylpyridine is neurotoxic to the nigrostriatal dopamine pathway. Nature 1986;319:56-57 25. Langston JW, Ballard P, Tetrud JW, Irwin I. Chronic parkinsonism in humans due to a product of meperidine synthesis. Science 1983;2 19:979-980 26. Burns RS, Chiueh CC, Markey SP, et al. A primate model of parkinsonism: selective destruction of dopaminergic neurons of the pars compacta of the substantia nigra by N-methyl-4phenyl-1,2,3,6-tetrahydropyridine. Proc Natl Acad Sci USA 1983;80:4546-4550 27. Slivka A, Mytilineou C, Cohen G. Histochemical evaluation of glutathione in brain. Brain Res 1987;409:275-284 28. Raps SP, Lai JCK, Hertz L, Cooper AJL. Glutathione is present in high concentrations in cultured astrocytes but not in cultured neurons. Brain Res 1989;493:398-401
Di Monte et al: GSH in PD SLl5