Redox Imbalance

In the Triad of Genetic Disturbances and Mitochondrial Dysfunction in Parkinson's Disease

Daniela Berg

INTRODUCTION: FREE RADICALS AND OXIDATIVE STRESS

The central nervous system is particularly vulnerable to oxidative stress (OS) as it utilizes large amounts of dioxygen but harbors relatively poor concentrations of antioxidants and related enzymes. Moreover, it contains a very high amount of polyunsaturated lipids, the most vulnerable biomacromolecule to oxidation (1). Free radicals in Parkinson's disease (PD) comprise mainly oxygen radicals (reactive oxygen species (ROS)) or oxides of nitrogen (reactive nitrogen species (RNS)). Moreover, transition metals like iron and copper contribute to the generation of OS as they have the ability to change oxidation numbers by one, allowing them to donate or accept single electrons. This ability makes them powerful catalysts of free-radical reactions.

ROS are generated as a result of normal metabolism. However, the deleterious condition termed OS occurs when ROS or RNS due to an excessive production overwhelms the protective defense mechanisms of a cell resulting in functional disruption and ultimately in cell death. The most important oxygen species in humans are H2O2, superoxide radical (O^-) and hydroxyl radical (HO*). Reactive nitrogen species include the radical nitric oxide (NO) and peroxynitrite (ONOO*-).

Although it is the neuronal loss of the substantia nigra (SN) that leads to many of the clinical symptoms of PD it is obvious that 30-40% of the changes in parameters of OS found in homogenates of nigral tissue cannot be restricted to these cells that account for only 1-2% of the whole nigral cell population (2). Changes also occur in other cell types, predominantly in glial cells, implying a concept of general metabolic failure in the SN of PD patients. The reason why it is the SN that is the target of the high degree of OS in PD may lay in its high energy metabolism and the high content of dopamine in its neuronal cells although dopaminergic cells are normally endowed with quite a number of protective mechanisms. Moreover, neuropathological studies have shown that primarily long fibers with scarce myelinization needing more energy degenerate (3). Lack of antioxidant capacities of these fibers makes them especially vulnerable to OS.

It is not yet clear, whether OS is a primary cause of neurodegeneration or a consequence of other pathogenetic factors. Studies in patients with incidental Lewy body disease (ILBD), which is supposed to represent presymptomatic PD, implicated that with

Oxidative Stress and Neurodegenerative Disorders Edited by G. Ali Qureshi and S. Hassan Parvez

© 2007 Elsevier B.V. All rights reserved.

the exception of reduced GSH levels there is no conclusive evidence of other markers of OS at an early stage of neurodegeneration (2). However, no final conclusion can be derived from this observation as only few samples have been examined with tissue homogenates rather than detailed studies on dopaminergic neurons. On the other hand, studies demonstrating oxidative dimer formation as the critical rate-limiting step for fibrillogenesis of a-synuclein provide an explanation that overproduction of ROS and/or impairment of cellular antioxidative mechanisms are primary events both in the initiation and in the progression of PD (4). It is therefore highly possible that OS may be both an initiator of neurodegeneration and a component of the pathogenetic process accelerating neuronal loss.

FACTORS CONTRIBUTING TO THE GENERATION OF OXIDATIVE STRESS (OS)

Dopamine

Formation of free radicals

Metabolism of dopamine produces ROS and might therefore account at least in part for the selective vulnerability of the SN pars compacta (SNc) in PD. Already during the process of dopamine synthesis, cytotoxic products like reactive dopamine quinone products may be formed (5). After synthesis, dopamine is directly taken up into synaptic vesicles. Here, dopamine is protected from oxidation by a very low pH that stabilizes the catechol structure and confers a milieu where protons are very strongly bound to oxygen atoms. The other mechanism protecting neurons from autooxidation of dopamine involves dopamine metabolism by monoamine oxidase (MAO). Autooxidation of dopamine leads to the production of dopaquinone and O2-. This reaction is catalyzed by metals, oxygen or enzymes like tyrosinase or xanthine oxidase. O2- is either metabolized into H2O2 or it reacts with nitric oxide, generating the strongly reactive peroxynitrite (ONOO*-). In the second step, dopaquinone is cyclisized to aminochrome. This may then be polymerized leading to the formation of neuromelanin (NM) or may be conjugated with GSH and reduced by one- or two-electron transfer catalyzed by quinone reductases (6). The conjugated aminochrome leukoaminochrome-GSH is very stable in contrast to unconju-gated aminochrome-reduced forms. Also, two-electron oxidation of aminochrome, which is catalyzed by DT-diaphorase is supposed to be neuroprotective, as the autooxidation rate of the produced o-hydroxychinone (leukoaminochrome) is very low. Therefore depletion of GSH or changes in the function of DT-diaphorase constitutes reduced cellular defense mechanisms leading to increased formation of ROS. It is generally assumed, that in PD autooxidation of dopamine may therefore be the consequence of an overproduction of dopamine, an inhibition or low expression of synaptic vesicle catecholamine transporters or inhibition or low expression of MAOs (6).

Dopamine linked to genetic defects and mitochondrial dysfunction

Besides the direct contribution to the generation of ROS, dopamine has been shown to form covalent oxidative adducts with a-synuclein leading to its retention in a protofibrillar form (7), which is capable of permeabilizing synthetic vesicles (8) enhancing dopamine leakage. In cultured human dopaminergic neurons, mutant a-synuclein has been shown to even trigger an elevation of cytosolic dopamine, enhancing dopamine-dependent toxicity (9).

Moreover, it has been shown that dopamine may inhibit complex I when injected into the brain ventricle of rats (10). The fact that rotenone only exerts its toxic influence on dopaminergic neurons in spite of inhibiting complex I throughout the whole brain implies that dopaminergic neurons are preferentially vulnerable to complex I defects (11).

Therefore, the selective vulnerability of dopaminergic neurons may be a result of dopamine-dependent OS as well as possible influence of defective proteolysis and mitochondrial dysfunction.

Neuromelanin (NM)

The large amount of NM in the SNc is unique to humans. It is generally regarded to be the result of the oxidation of dopamine and noradrenaline (12,13). This, however, has been questioned due to the fact that not all dopaminergic neurons of the SN contain NM and long-term l-DOPA treatment does not seem to enhance NM concentration in surviving neurons. As in PD primarily NM-containing neurons degenerate (14), with the largest pigmented neurons being preferentially lost, a cytotoxic effect of NM contributing to OS has been proposed. Conversely, the less pigmented ventral tier of the SN is the first to degenerate in PD (15). NM is an excellent chelator of metal ions, especially iron (16,17), therefore, a neuroprotective role of NM is discussed (18,19). Iron bound to NM accounts for 10-20% of the total iron in the SN in normal subjects aged 70-90 years (20,21). In PD, however, the absolute concentration of NM within the SN is dramatically decreased. The level of redox activity detected in NM aggregates, however, was found to be substantially increased in PD patients the highest being in patients with the severest neuronal loss (22). It has been supposed that the amount of iron determines the role of NM: In the situation of normal iron levels, this redox-active metal is sequestered. In the presence of excess iron, however, NM promotes the formation of ROS and fosters the release of iron into the cytoplasm (23-25). In accordance with this hypothesis, the pigment isolated from patients with PD showed a lower total magnetization than control NM suggesting a progressive migration to the cytosol (26). Additionally, NM can bind a variety of potentially toxic substances like MPP+, the neurotoxic metabolite of MPTP or pesticides suggesting a contribution to neurotoxin-mediated neurodegeneration (19-27). It can therefore be hypothesized that not NM itself but rather its interaction with iron, catechols and neurotoxic metabolites may account for its contribution to OS.

Only recently a possible radical cross-linking between the polycatecholic framework of NM and the isoprenoid chain of dolichol, a lipid component of intact NM granules has been described (28,29). In the NM extracted from PD brains, the whole NM pigment appeared to be mainly composed of highly cross-linked, protease-resistant lipoproteic material (30). It has been hypothesized that a-synuclein could localize preferentially within the NM lipid phase, thereby facilitating the interaction between these pathologically important substances (29-31).

Transition metals

Transition metals are essential in most biological reactions, e.g. for the synthesis of a great number of enzymes, in the synthesis of DNA and RNA, in O2 transport and a number of redox reactions. However, by their ability to undergo one-electron transfer, they are also potentially dangerous, enabling autooxidation (e.g. dopamine and ascorbate), conversion of H2O2 to HO* or decomposition of lipid peroxides to reactive peroxyl and alkoxyl radicals. Besides the "free" metal ions, metal ions contained in proteins may be catalytic like the iron bound to heme. A careful regulation of cellular balance is therefore essential.

Iron

The content of iron, which is essential for many biological processes including its role as a cofactor for the synthesis of dopamine (32), is, under physiologic conditions, higher in the basal ganglia and SN than in most other regions of the brain (33). In PD, iron content of the SNc is additionally about 35% elevated. However, it is not the increase in total iron that implicates OS as long as a concomitant increase in proteins keeps it stored in a redox inert form. In PD, an increase of the Fe(III):Fe(II) ratio from 2:1 to almost 1:2 has been found (34-37). An important site of iron release is microglia. Here, superoxide and a number of oxidized catechols may lead to the release of iron from ferritin (38) thereby contributing to free-radical-induced cell damage.

Increased levels of iron and Fe(II) enhance the conversion of H2O2 to *OH via the Fenton reaction and favor a greater turnover in the Haber-Weiss cycle, which leads to an amplification of OS (39). On the other hand, OS may increase the levels of free iron. The mechanisms include the release of iron from ferritin by O2-, from heme proteins like hemoglobin and cytochrome c by peroxides and from iron-sulfur proteins by ONOO*- (40).

Besides the contribution to the formation of highly ROS, iron has been shown to interact with a-synuclein (41-43) enhancing the conversion of unfolded or a-helical conformation of a-synuclein to ^-pleated sheet conformation, the primary form in Lewy bodies (LBs). Colocalization of proteins involved in brain iron metabolism and LBs (44,45) is a further implication for the involvement of iron in the neurodegenerative process in PD. It is not entirely clear yet, at what time in the pathophysiological cascade of PD iron accumulation occurs. Iron accumulation induced by toxin-mediated neurodegeneration in animal models suggests it to be a secondary phenomenon (46,47). However, high iron diet, fed to weanling mice, has been shown to lead to marked reduction of SN glutathione levels, a finding known to occur very early in PD (48,49). Data from recent transcranial ultrasound studies also imply iron accumulation to occur very early in the disease process constituting rather a primary cause of the disease in idiopathic PD (50-52). In contrast, in patients with monogenetic PD the ultrasound finding indicates less iron accumulation (Schweitzer et al., Neurol, in press). Therefore, it may be possible that iron contributes to different degrees at different stages to the pathophysiological cascade of PD. In idiopathic PD, a more causative role earlier in the disease process may be postulated, while in mono-genetically caused PD other factors may have greater influences on disease development and progression. Interestingly in single cases of apparently "idiopathic" PD, an association of sequence variations in some genes encoding for iron-metabolizing proteins within the brain and PD has been established (45-53) while such an association could be ruled out in others (54,55).

Copper

Copper is on the one hand, essential for the function of key metabolic enzymes but may enhance production of ROS when it reacts uncontrollably on the other hand. These reactions may even be aggravated under conditions of OS as exposure to ONOO*- may lead to the release of copper from ceruloplasmin. Dopaminergic neurons are especially vulnerable as copper neurotoxicity seems to depend on dopamine-mediated copper uptake (56). Also, exogeneously acquired increased copper levels may be deleterious as shown by a population-based case-control study which provided evidence that chronic occupational exposure to copper is associated with PD (57). Moreover, dietary and pharmacological manipulations of copper modify the course of the disease in mouse models of PD in ways that suggest a role for this metal in disease pathogenesis (58). It has been shown that copper may accelerate aggregation of a-synuclein to form fibrils (59) and that copper-mediated stress is linked to mitochondrial dysfunction as a result of decreased activity of cytochrome c oxidase (60).

Manganese

This metal is essential for a number of enzymes including SODs, arginase, hydrolase and carboxylase enzymes. However, chronic exposure to even moderate amounts of manganese over longer periods of time may induce parkinsonism similar to idiopathic PD. Similarly, combination of high intake of iron and manganese has been found to be related to PD (61). The main pathophysiological impact of manganese is supposed to be the promotion of rapid dopamine oxidation in the brain leading to severe destruction of brain tissue at the striatum and pallidum (34,62,63). Moreover, it may incur depletion of levels of peroxidase and catalase (64). In vitro and animal models suggest that manganese directly inhibits mitochondrial function preferentially by inhibiting mitochondrial complex III (65). Additionally, manganese has been shown to accelerate a-synuclein fibril formation (66).

Zinc and magnesium

Not all metals enhance production of ROS. Zinc, which on the one hand may contribute to the generation of OS by interference with the mitochondrial complex I, may act on the other hand as an antioxidant by displacing iron ions from their binding sites and inhibiting iron-dependent radical reactions. It exerts this influence by binding to thiol groups, inhibiting nitric oxide synthase and inducing Zn2+-containing, antioxidative proteins (67,68). Accordingly, patients with PD showed a significantly decreased zinc status established by a zinc tally test (69) and also the CSF levels of zinc were significantly decreased in PD patients as compared with controls (70). Similarly, deficiency of magnesium, a cofactor for multiple enzymes, may increase oxidative damage. Moreover, in vitro studies have shown that magnesium may inhibit the aggregation of a-synuclein induced either spontaneously or by incubation with iron (71).

INTERACTION OF ROS WITH GENETIC INFLUENCES, PROTEIN AGGREGATION AND LEWY BODIES (LBS)

Sporadic PD mimics monogenetic PD due to an interaction of ROS/RNS with the respective genes

Normally, abnormal proteins are ubiquitinated and degraded by the ubiquitin-proteasome system (UPS). The UPS holds a unique role in intracellular metabolism as it represents the major route of protein degradation and seems to be specifically regulated at multiple levels. Genes encoding for proteins involved in the monogenetic forms of PD directly link the pathogenesis of PD to the UPS. One of the key proteins is parkin, an E3 ubiquitin ligase, adding ubiquitin to specific substrates thereby marking them for degradation by the proteasome. Moreover, ubiquitin carboxy-terminal hydrolase-Ll (UCH-L1), a deubiquiti-nating enzyme recycling ubiquitin, links aberrant UPS activity and PD. Also, mutations in a-synuclein, DJ1 and PINK1 may contribute to UPS dysfunction. Mutations in the parkin gene have been shown to increase the formation of ROS and RNS (2). On the one hand, there is increasing evidence that nitrosative or oxidative stress results in the malfunction of proteins involved in the proper function of the UPS like parkin or UCH-L1 (72-75). In the brains of patients with PD as well as in animal models of the disease, parkin was found to be S-nitrosylated (73-75). S-nitrosylation of parkin was observed to stimulate its E3 ligase activity resulting in an initial increase in enzyme activity, leading to autoubiq-uitination with subsequent decrease in activity (75,76). It has been hypothesized that the initial increase of parkin activity could contribute to LB formation (76,77) whereas the subsequent decrease could lead to UPS dysfunction.

Similarly, it was found that oxidation of UCH-L1 results in a loss of its important hydrolase activity (72) probably contributing to UPS dysfunction. Also, this protein was detected in an oxidized form in sporadic PD brains.

Also, DJ1 is very susceptible to oxidation (78). DJ1 has been proposed to be a redox-dependent chaperone that inhibits protein aggregation in vitro and intracytoplas-mic inclusions in vivo (79). Thus, enhanced OS may lead to a loss of its secondary structure resulting in a loss of its ability to inhibit a-synuclein fibrillation.

Moreover, in vitro studies showed that aggregation of a-synuclein, as seen in PD, may be caused or catalyzed by exactly those factors that had already been suspected as risk factors for PD such as certain pesticides, heavy metals, advanced glycation endproducts (AGEs) and others leading to cellular OS (80,81). As the oxidative dimer formation is supposed to be the rate-limiting step for fibrillogenesis, overproduction of ROS and/or impairment of cellular antioxidative mechanisms may be regarded as primary events in the initiation and progression of PD (4).

It is therefore highly possible that altered activity of proteins involved in monogenetic PD like parkin, UCH-L1 and DJ1 as well as increased fibrillation and aggregation of a-synuclein induced by ROS/RNS may be critically involved in the etiology of sporadic PD.

UPS function and oxidative/nitrosative stress

Proper function of the proteasome becomes even more important under conditions of OS, when proteins are oxidized. However, OS can impair the ubiquitin-proteasome directly and products of OS can damage the 26S proteasome (2), particularly when GSH levels are reduced (82). Moreover, complex I inhibition has been shown to decrease proteasomal activity, thereby increasing neuronal vulnerability to normally subtoxic levels of free radicals (83). The result is an accumulation of oxidized proteins, which again can generate free radicals contributing to cell toxicity and which are more prone to aggregation (84).

In PD brain, areas with the highest levels of a-synuclein are those associated with LBs. It has therefore been speculated that LBs result from altered handling of oxidized proteins and may at least initially represent a protective mechanism of the cell from the toxicity of protein accumulation (48). Also, the sequestration of toxic iron by LBs indicates a protective role of these inclusions (85). Some investigations, however, indicate that LBs might contribute to the pathological cascade (86,87).

Investigation of the role of ROS and protein metabolism therefore shows a complex, inevitably linked and integrated interaction including endogenous production of OS, chronic exposure to environmental agents and variations in genes impairing the UPS (88,89). It seems that an individual combination of these factors may lead to neurodegeneration with the clinical picture of sporadic PD.

0 0

Post a comment