X-linked adrenoleukodystrophy (X-ALD) is an inherited neurodegenerative disorder expressed as four disease variants characterized by adrenal insufficiency and graded damage in the nervous system. X-ALD is caused by a loss of function of the peroxisomal ABCD1 fatty-acid transporter, resulting in the accumulation of very long chain fatty acids (VLCFA) in the organs and plasma, which have potentially toxic effects in CNS and adrenal glands. We have recently shown that treatment with a combination of antioxidants containing α-tocopherol, N-acetyl-cysteine and α-lipoic acid reversed oxidative damage and energetic failure, together with the axonal degeneration and locomotor impairment displayed by Abcd1 null mice, the animal model of X-ALD. This is the first direct demonstration that oxidative stress, which is a hallmark not only of X-ALD, but also of other neurodegenerative processes, such as Alzheimer’s disease (AD), Parkinson’s disease (PD) and Huntington’s disease (HD), contributes to axonal damage. The purpose of this review is, first, to discuss the molecular and cellular underpinnings of VLCFA-induced oxidative stress, and how it interacts with energy metabolism and/or inflammation to generate a complex syndrome wherein multiple factors are contributing. Particular attention will be paid to the dysregulation of redox homeostasis by the interplay between peroxisomes and mitochondria. Second, we will extend this analysis to the aforementioned neurodegenerative diseases with the aim of defining differences as well as the existence of a core pathogenic mechanism that would justify the exchange of therapeutic opportunities among these pathologies. This article is part of a Special Issue entitled: Metabolic functions and biogenesis of peroxisomes in health and disease.


► Oxidative and metabolic damage contribute to axonal degeneration in X-ALD. ► We underscore the commonalities between X-ALD and major neurodegenerative diseases. ► We focus on therapeutic approaches against oxidative damage for X-ALD.

1. Introduction

X-linked adrenoleukodystrophy (X-ALD: McKusick no. 300100) is a neurometabolic genetic disorder characterized by progressive demyelination in the central nervous system (CNS), axonopathy in the spinal cord, and adrenal insufficiency. The disease is caused by mutations in the ABCD1 (ALD) gene in Xq28, that encodes for the peroxisomal ABCD (ALD protein or ALDP) transporter [1], which imports VLCFA-CoA esters into the peroxisome, where they are degraded by β-oxidation [2]. X-ALD is the most common monogenic leukodystrophy and peroxisomal disorder, occurring in at least 1 out of 17,000 males [3], [4].

Four major disease variants have been described: i) a late-onset and slowly progressing form affecting adult men and women, called adrenomyeloneuropathy (AMN), as it presents peripheral neuropathy and distal axonopathy in spinal cord, often, but not always, associated with axonal or demyelinating peripheral neuropathy; ii) AMN with cerebral demyelination in adult males that becomes inflammatory in many cases; iii) cerebral inflammatory demyelination in boys and adolescent males; iv) and, finally, peripheral adrenal insufficiency (Addison’s disease) in boys, adolescents and adult males which is not fully penetrant. Patients with this final form are at, however, a 100% risk of developing later cerebral demyelination or AMN. The cerebral inflammatory demyelinating forms of X-ALD that occur in boys, adolescent and adult males are fatal.

It is not well understood why X-ALD either affects the spinal cord, the peripheral nerves, the brain white matter or the adrenal glands. The observation that approximately 20% of male patients with adult AMN develop cAMN [5], suggests the possibility of a pathogenic continuum of increased severity.

Whatever its location or variant, X-ALD is a disease that affects axons. In the brain variants demyelination causes severe axon damage and neuronal death, while in AMN there is direct damage to axons in the spinal cord and in peripheral nerves, with demyelination being secondary. In this review we will discuss: i) how the abnormal accumulation of VLCFA may lead to either form of X-ALD by creating a syndrome of energy metabolism impairment and inflammation via oxidative stress, and ii) whether some of these lessons can be applied to major neurodegenerative diseases, in order to better understand the pathogenic interplay of staple disease factors such as oxidative stress, inflammation and energy metabolism disturbance.

2. A word on oxidative stress and damage

Chemical reactions in nervous system are under strict enzyme control and conform to a tightly regulated metabolic program in order to minimize unnecessary side reactions. Nevertheless, apparently uncontrolled and potentially deleterious reactions occur, even under physiological conditions. The term “reactive oxygen species” (ROS) comprises a variety of molecules and free radicals (chemical species with one unpaired electron) physiologically generated from the metabolism of molecular oxygen [6]. ROS include superoxide anion (O2), hydrogen peroxide (H2O2), hydroxyl radical (HO), peroxyl radical (RO2), alkosyl radical (RO), hydroperoxyl radical (HO2), hypochlorous acid (HOCl), hypobromous acid (HOBr), and singlet oxygen (O2) [6].

ROS are extremely reactive and have damaging effects. Despite the fact that ROS can be generated at various sites and as a consequence of various conditions, including ischemia–reperfusion and enzymatic reactions (e.g. the membrane NADPH oxidase, lipoxygenases, cyclooxygenases, peroxidases, and other heme proteins, the enzyme xanthine oxidase, β-oxidation in peroxisomes, and detoxifying reactions in hepatic P-450 microsomes) in healthy cells under physiological conditions, most ROS originate in mitochondria [6]. Superoxide anion, the product of a one-electron reduction of oxygen, is the precursor of most ROS, and a mediator in oxidative chain reactions. The character of the radical is not limited to oxygen containing species, as nitrogen-, chloride- and sulfide-containing molecules are also radical species that could also play significant pathophysiological roles. Globally, in cells of the nervous system the major sites of physiological ROS generation are the complex I and III of the mitochondrial electron transport chain, which contain several redox centers (flavins, iron–sulfur clusters, and ubisemiquinone) capable of transferring one electron to oxygen to form a superoxide anion [7].

Since nervous cells continuously produce free radicals, their oxidative stress homeostasis is only guaranteed if an adequate pool of endogenous cellular antioxidants is present. A large battery of antioxidant defenses, both enzymatic and nonenzymatic, have been selected and conserved during evolution [8]. Superoxide dismutase (SOD) eliminates the superoxide radical converting it to oxygen and H2O2. There are different forms of this enzyme: a Cu,Zn form in the cytosol and in the intermembrane mitochondrial compartment, a Mn form in the mitochondrial matrix, and another form in the extracellular compartment. The mitochondrial enzyme is essential for life, its decline leading to mitochondrial dysfunction, pathology, or neonatal lethality, depending on the level of depletion; MnSOD overexpression can protect against pro-apoptotic or pro-ischaemic insults. SOD alone, however, cannot be considered strictly as an antioxidant because, although it eliminates superoxide radicals, it produces another molecule which could be a ROS source, H2O2. Thus, other enzymes work in a coordinated manner, eliminating the hydrogen peroxide produced by SOD and other sources. Two main kinds of enzymes perform this task. Catalase decomposes H2O2 at high rates but shows low affinity for peroxide and is most useful during peaks of H2O2 production or accumulation. These peaks must occur in vivo, because acatalasemia increases oxidative stress and induces pathologies in humans [9]. Glutathione peroxidases (GSH-Px), present in selenium- and nonselenium-dependent forms, are complementary to catalase, since they decompose H2O2 slowly but with higher affinity. Thus, they are most useful to decompose the small amounts of peroxide continuously produced inside cells. Different GSH-Px forms have been described in the cytosol, mitochondria and cellular membranes, and they can reduce inorganic or organic peroxides. At least five selenium-containing GSH-Px have been identified, and their activity can be manipulated by changing dietary selenium levels. These enzymes use the reduced form of glutathione (GSH) to decompose peroxide. In this process oxidized glutathione (GSSG), which is highly toxic to cells, is generated. GSSG is then reduced back to GSH by glutathione reductase (GSH-Red), another antioxidant enzyme, which uses the NADPH generated by the pentose phosphate pathway or by the NADH–NADPH–transdehydrogenases. Indeed, GSH is a cofactor of GSH-Px and the major endogenous antioxidant produced by the cells, participating directly in the neutralization of free radicals and reactive oxygen compounds, as well as maintaining exogenous antioxidants such as vitamins C and E in their reduced (active) forms. Glutathione exists in reduced (GSH) and oxidized (GSSG) states, with GSH being predominant and thus maintaining the cytosol in strong “reducing” conditions. Another thiol-related redox-active substance is the protein thioredoxin. Thioredoxin has a redox-active disulfide/dithiol at the active site. Its activation by ROS regulates transcription factors (such as nuclear factor kappaB and AP-1), thus mounting a defensive response. Accordingly, it is induced by various oxidative stresses and is translocated to the nucleus. Thioredoxin is cytoprotective against oxidative stress by scavenging ROS in cooperation with peroxiredoxin/thioredoxin-dependent peroxidase.

Apart from GSH, vitamin C is the next most abundant reduced non-enzymatic antioxidant inside cells [10]. It is endogenously synthesized in most vertebrates (although not in human beings, fruit bats or guinea pigs), and it is maintained at levels as high as 1 mM in some tissues. After reacting with ROS, the oxidized form of ascorbate must be reduced by NADPH-, GSH- or NADH-dependent reductases to regain antioxidant capacity [10].

Tocopherols and carotenoids are the main radical scavenger antioxidants that act in lipophilic environments of cells, a fact that is particularly relevant for the nervous system. The major scavenger inside membranes is d-α-tocopherol (vitamin E) [10]. Most membranes are thought to contain approximately one tocopherol molecule per thousand lipid molecules. Vitamin E acts on lipid peroxyl groups inside membrane bilayers, reducing them to hydroperoxides, and thus inhibiting the propagation of the peroxidative chain reaction. It breaks the chain reaction of lipid peroxidation but is itself converted to a radical during the process. Vitamin E also reduces lipid alcoxyl radicals to lipid alcohols. Oxidized vitamin E can be recycled back to its reduced form by ascorbate or ubiquinone (coenzyme Q). Other lipophilic antioxidants comprise the carotenoids and ubiquinol. The former quench singlet oxygen, and interact with other ROS at physiological tissue oxygen partial pressures. Ubiquinol, the reduced form of coenzyme Q, is also an important antioxidant. It is a hydroquinone that is synthesized and present in all cellular membranes, and its antioxidant activity is exhibited through scavenging of lipid radicals and reduction of vitamin E radical. Regeneration of coenzyme Q is performed by reductases that use NADPH or NADH as cofactors.

ROS also participate in signal transduction in physiological conditions [11], but when net ROS production surpasses the above-mentioned antioxidant barriers, a condition of “oxidative stress” arises in which ROS may cause structural and functional changes in all cellular constituents including DNA, RNA, lipids and proteins. Consequently, oxidative damage (the term used to cover the chemical modification induced in biological molecules by free radicals if such change results in function impairment) occurs whenever the ROS produced by mitochondria avoid detoxification, the steady-state level of molecular oxidative damage depending on the relative rates of damage accumulation, repair, and degradation. The primary cellular target of oxidative stress depends upon the cell type, the nature of the stress imposed, the susceptibility to oxidation of the target molecule, the site of generation, the proximity of ROS to a specific target, and the severity of the stress. In this context, protein oxidation demands a special mention because proteins constitute the major ‘working force’ for all forms of biological work. Furthermore, their exact conformation and pattern of folding are tightly related to their activity and function. So, the resulting loss of function and structural integrity of oxidatively modified proteins can have a wide range of downstream functional consequences and may be the cause of subsequent cellular dysfunction and tissue damage. In this scenario, the primary change of metal-catalyzed protein oxidation is carbonylation represented in the carbonyl derivatives glutamic semialdehyde (GSA) and aminoadipic semialdehyde (AASA) [12], [13], which are among the most specific markers of protein oxidation. In addition, oxidative modifications of proteins may be caused by reactive carbonyl compounds derived from the oxidation of carbohydrates and lipids, leading to the formation of advanced glycation and lipoxidation end-products in proteins including malondialdehyde-lysine (MDAL), carboxylmethyl-lysine (CML), and carboxyethyl-lysine (CEL) [14], [15].

The high content of highly peroxidizable polyunsaturated fatty acids in membranes, the presence of catecholamines prone to oxidation, the elevated oxygen consumption, and the relatively poor expression of enzymatic antioxidant defenses in the CNS, compared to other tissues, help explain its high susceptibility to oxidative damage [7], [16]. An often overlooked fact is, however, worth stressing: The unequivocal diagnosis of oxidative damage requires both detection of ROS-mediated structural changes and proof of functional impairment, for the physiological actions of ROS involve oxidation of molecules, too [11].

The double-edged role of oxidation is exemplified in nitric oxide (NO), which is the result of oxidation of l-arginine by nitric oxide synthases (NOS), which plays a key role in host defense, vascular reactivity and neurotransmission. However, NO is the source of very deleterious reactive nitrogen species (RNS) like peroxynitrite. RNS are produced by highly activated cells of the immune system, including astrocytes and microglia, and characteristically lead to the formation of 3-nitrotyrosine as well as to the oxidation of some amino acid residues [17]. Glutathione-S-transferase, which uses GSH as a substrate, destroys peroxynitrates [18]. Thus, antioxidant reactions aimed at increasing GSH also protect the cell from RNS.

3. Oxidative stress and damage in X-ALD patients

Signs of oxidative modification have been reported in postmortem X-ALD brains as well as in skin-derived fibroblasts, plasma, and blood cells, which are surrogates to study diseases of the CNS in which affected tissues can be accessed only post-mortem. Thus, there is a very robust increase in nitrotyrosylated proteins, NOS2 and lipid peroxidation products within the neuropathological inflammatory lesion of cALD or cAMN, associated with astrocytosis and microgliosis [19], [20]. Manganese superoxide dismutase (MnSOD), a mitochondrial enzyme that increases in response to oxidative stress, was also upregulated as determined by immunohistochemistry, pointing to mitochondrial alterations in the pathological zones [20]. Interestingly, while oxidation markers were characteristically present in the core of cerebral and cerebellar white matter lesions, these were also detected beyond [20]. These data are of relevance because they provide a picture of early stages in disease progression that are plausibly more amenable to therapeutic intervention. Recently, we have confirmed by a functional genomic analysis the existence of a proinflammatory reaction in still normal looking white matter of cALD and cAMN patients [21].

In plasma, Vargas’ laboratory has reported increased lipid peroxidation by analysis of TBARS (thiobarbituric acid reactive substances) in plasma of cALD and AMN patients, and, to a lesser extent, in asymptomatic carriers of ABCD1 mutations [22]. The antioxidant defense, measured by the capacity to quench peroxidase, inversely mirrored these changes as it was reduced in cALD and AMN patients, while appearing normal in asymptomatic patients. This suggests that an efficient anti-oxidant defense could impede the disease development in asymptomatic subjects [22]. There is also increased oxidative stress in X-ALD fibroblasts [23], and free radical production in X-ALD lymphoblasts [24]. This implies that oxidative stress is a generalized phenomenon in different cell types linked to the loss of function of the ABCD1 transporter/ALD protein and the accumulation of fatty acids. Moreover, our laboratory has shown by very sensitive gas chromatography/mass spectrometry (GC/MS) techniques that the markers MDAL, CEL and CML, GSA and AASA were increased by two-fold in fibroblasts derived from X-ALD patients [25]. The observations were extended with analysis of the effects of exogenously-added C26:0, which caused ROS and oxidative damage in proteins. It is worth stressing that X-ALD fibroblasts responded more acutely to C26:0 than did control cells, suggesting impaired anti-oxidant defense consistent with the detected depletion of GSH [25].

4. Early oxidative stress and damage in X-ALD mice

The advantage of a representative mouse model over postmortem patient materials is the possibility to carry out a longitudinal characterization of the disease process, thus allowing the discrimination of causative events from epiphenomena. Three mouse models for X-ALD have been independently generated following a classical strategy of knocking out the Abcd1 gene, located also in the X chromosome in the mouse genome [26], [27], [28]. None of the mouse mutants present cerebral demyelination or inflammatory signs up to six months of age, in spite of elevated levels of VLCFA in the nervous tissue, adrenal glands and other organs [26], [27], [28]. However, histological signs of axonopathy in sciatic nerves are detected at 16 months of age, and the animals display microglia activation and astrocytosis and axonal swellings and damage at 20–22 months of age, concomitant with locomotor alterations and altered motor nerve conduction velocities [29], [30]. Thus, this mouse model mimics the spinal cord axonopathy present in patients affected with pure AMN [29], and constitutes a bona fide model to dissecting physiopathogenetic mechanisms.

Regarding oxidative homeostasis, no evidence of ROS production or oxidative modification was reported in brain, adrenal cortex or kidney from 3-month-old Abcd1 null mice [20]. However, there were increases in MnSOD immunostaining in cerebellum, liver, and adrenal and kidney cortices [20], pointing to an incipient redox alteration, because MnSOD is characteristically activated by oxidative stress. We thus set out to thoroughly investigate the presence and time course of oxidative damage in the nervous system, using GC/MS to quantify oxidation-derived protein modifications such as the above explained GSA, AASA, CML, CEL and MDAL. While no increase in the concentration of markers of oxidative modification was observed at 18 days, MDAL was increased at 3.5 months, and GSA, AASA, CEL and MDAL at 12 months, in a selective manner, as no increased levels were observed in cerebral cortex or liver. Thus, oxidative stress occurs well before disease onset in X-ALD mice, increases with time, and is target-organ specific. Lipid peroxidation as evidenced by MDAL arises as the earliest quantitative disease marker [25]. Of note, the same quantitative markers were increased by two-fold in fibroblasts derived from X-ALD patients.

In an ensuing study we discovered by redox proteomics [31] that the oxidative modification of proteins specifically affects five key enzymes of glycolysis and tricarboxylic acid cycle (TCA): aldolase A, phosphoglycerate kinase, pyruvate kinase, dihydrolipoamide dehydrogenase and mitochondrial aconitase. Moreover, we confirmed that the oxidative alteration led to damage because we discovered, by metabolomic measurements of substrates and/or products, that these enzymes were inactivated, and we showed depletion of NADH and ATP [32]. These data confirm the notion hinted at in our previous studies: there is a metabolic derangement in spinal cords of Abcd1 null mice prior to development of neurological symptoms, thus lending support to the idea that energy failure may be a pathogenic factor, at least in adult forms of X-ALD (see below).

5. What causes oxidative stress in X-ALD?

The studies from our laboratory in human X-ALD fibroblasts showing that hexacosanoic acid in excess directly increases steady-state ROS production, depletes GSH, and decreases mitochondria membrane potential serve to argue that VLCFA is the agent initiating oxidative stress in X-ALD [25]. This is not an effect restricted to hexacosanoic acid, as an excess of the monounsaturated C26:1 fatty acid, also a VLCFA whose levels are increased in X-ALD, generates ROS in human fibroblasts [33].

By what mechanisms does VLCFA generate free radicals? As discussed above, mitochondria, peroxisomes, endoplasmic reticulum, microsomes, nucleus and plasma membrane oxidases are potential sources of ROS. In X-ALD, neither the exact source of free radicals nor the molecular mechanism of production has yet been elucidated; this information is essential to the development of tailored therapeutic strategies. In light of recent findings, several scenarios are discussed below. We highlight mitochondria as the most important ROS-producing organelle as a result of electron-transfer reactions, and also as relevant to the subject of this review, peroxisomes, which carry β-oxidation of fatty acids. Peroxisomes are an important source of total cellular H2O2 production [34]. They contain a number of H2O2-generating enzymes including glycolate oxidase, d-amino acid oxidase, urate oxidase, l-α-hydroxyacid oxidase, and fatty acyl-CoA oxidase. Peroxisomal catalase utilizes H2O2 produced by these oxidases to oxidize fatty acids in “peroxidative” reactions [35].

5.1. The peroxisome–mitochondria connection

There is growing evidence that both organelles exhibit a closer interrelationship than previously noted. This cross-talk includes: i) metabolic cooperation, for instance, the mitochondria can oxidize hexacosanoic CoA produced in peroxisomes by a CoA synthase [36], indicating that VLCFA derivatives can be shuttled from peroxisomes to mitochondria for full oxidation; ii) vesicular trafficking through Vps35 [37]; iii) a shared biogenesis through the activation of peroxisome proliferator activated receptors/peroxisome proliferator gamma coactivator-1 (PPAR/PGC-1) transcription factors [38] and a key component of the fission machinery, the Drp1 protein [39], [40], and iv) a redox connection. Thus, Koepke et al. [41] have observed that inhibition of peroxisomal catalase not only leads to an H2O2 elevation, but that it also causes mitochondrial dysfunction, as the treated cells show decreased inner mitochondrial membrane potential concomitant with enhanced mitochondrial ROS production. These findings have recently been reinforced in an elegant study showing that generating ROS inside peroxisomes disturbs the mitochondrial redox balance, which may lead to excessive mitochondrial fragmentation and ROS production [42]. The close relationship between both organelles is finally supported by morphological evidence of mitochondria impairment in peroxisomal Pex5 knockout mutant mice [43] and Zellweger syndrome patients, who lack functional peroxisomes [44]. Likewise, Abcd1 and Abcd2 knockout mice present abnormal, swollen mitochondria filled with lipidic inclusions and condensation of cristae in neurons of spinal cord [45], suggesting a role for the mitochondria in the physiopathology of peroxisomal disorders.

An excess of VLCFA may also affect mitochondria directly. As noted, mitochondria are major sources of ROS in the CNS. They contain redox carriers that can transfer single electrons to oxygen, thus generating the superoxide radical. Enzymes of the tricarboxylic acid cycle such as aconitase and ketoglutarate dehydrogenase, as well as the electron transport chain (complexes I, II, and II) and monoamine oxidases are among the mitochondrial redox carriers also generating superoxide radicals. As noted above, mitochondria also contain enzymes able to detoxify ROS. Superoxide is transformed into hydrogen peroxide by superoxide dismutases (SOD), enzymes which work in conjunction with catalases and glutathione peroxidases to remove H2O2 from mitochondria. It is worth noting that SOD1 and SOD2 have reduced expression in spinal cords of Abcd1 null mice [25].

Interestingly, in isolated mitochondria, long-chain fatty acids such as the branched chain phytanic acid stimulate ROS production by inhibition of the forward electron transport in the respiratory chain [46], [47]. Similar experiments performed with the VLCFA C22:0, C24:0 and C26:0 showed that mitochondria exposed to C22:0 and C24:0 fatty acids, but not C26:0, suffered uncoupling and inhibition of the respiratory chain. By contrast, while mitochondrial membrane potential was markedly decreased, in particular by C22:0, in cultures of oligodendrocytes, neurons and astrocytes, none of the VLCFA used stimulated ROS production from isolated mitochondria. Instead, the highly toxic effects of C26:0 were attributed to dysfunction of calcium homeostasis [48]. This is a canonical calcium- and mitochondria-mediated apoptosis. A caveat of this study is the time frame of cellular demise. In the CNS, X-ALD is characterized by slow degeneration and ultimately selective cellular death over years, even in the severe childhood demyelinating variant, rather than by acute cell death within hours, suggesting that other pathogenic mechanisms play roles in the detrimental effects of VLCFA in a real physiopathological setting.

In addition, it should be noted that experiments were performed in whole cells or isolated mitochondria from rat brain or liver with intact Abcd1 function. Mitochondrial function, and its response to VLCFA load, should be addressed in mitochondria from Abcd1 mouse models or human cells. In addition, studies of VLCFA cytotoxicity have mostly been performed using free fatty acids. These procedures might bring a limitation in, as the largest pool of fatty acids in the cells is not found free, but as components of complex lipids. In particular, VLCFA are constitutive of phospholipids and other lipidic species, such as gangliosides, phosphatidylcholine, and cholesterol ester fractions of brain myelin, as well as in the proteolipid fraction [49], [50], [51].

Finally, the recently uncovered loss of function of key mitochondrial enzymes of the Krebs cycle, aconitase and α-ketoglutarate dehydrogenase in X-ALD [32], supports a functional impairment of mitochondria. Note that these enzymes are targets of oxidative damage in spinal cords of Abcd1 null mice. That is, a vicious cycle arises whereby the mitochondrial enzymes are at the same time targets of free radicals, and ROS contributors when they are impaired.

5.2. Secondary peroxisomal dysfunction

ABCD1 loss results not only in accumulation of VLCFA outside peroxisomes, but also in another defective peroxisomal function, the synthesis of plasmalogens. These vinyl ether phospholipids with reported antioxidant activities are decreased in brains from X-ALD patients [52], and lower levels of plasmalogens have been reported to exacerbate pathology in the Abcd1 null mouse [53]. However, it is not clear how ABCD1 disruption may reduce plasmalogen contents. On the one hand, peroxisome biosynthesis appears to require acyl-CoA generated by the peroxisomal β-oxidation pathway [54], suggesting that a lower import of the VLCFA-CoAs, the substrate of β-oxidation, may decrease plasmalogen biosynthesis by reducing acyl-CoA availability. But, on the other hand, mice with a severe defect in peroxisomal β-oxidation, the MFP2 knockout mice, have normal levels of plasmalogens in brain [55], indicating that plasmalogen synthesis can proceed in the absence of β-oxidation. Whatever the reason underlying the decrease, the reduced contents of plasmalogens, together with the lower contents of the antioxidant enzymes SOD1 and SOD2 [25], may certainly render the nervous tissue more vulnerable to free radical attacks. Accordingly, loss of Abcd1 gene function sensitizes fibroblasts to death upon GSH depletion in the culture medium. Overall, the data indicate that Abcd1 dysfunction hampers oxidative stress homeostasis by increasing ROS while decreasing antioxidant defenses. This hypothesis is reinforced by the fact that selective peroxisome ablation in oligodendrocytes leads to cerebral axonal degeneration and inflammatory demyelination in brain, suggesting that the contribution of peroxisomal-related additional metabolic derangements is necessary to ignite the inflammatory response in mouse brain, and perhaps in X-ALD patients [56], [57].

5.3. Cellular membranes

NADPH oxidase is a plasma membrane-bound complex that generates superoxide by transferring electrons from NADPH inside the cell across the membrane, and coupling these to molecular oxygen. Superoxide can spontaneously form peroxide. In the brain, NADPH oxidase is found in astrocytes and microglia [58], [59]. Fatty acids have been described as NAPDH oxidase activators in neutrophils [60] and epithelial cells [61]. Indeed NADPH oxidase has been shown to generate free radicals in fibroblasts from X-ALD patients, which do accumulate VLCFA, concomitant to an increase in the protein contents of its membrane anchored catalytic subunit p91PHOX, while mRNA levels remained normal, as do the protein contents of the cytoplasmic NADPH subunits p47PHOX or p67PHOX [24]. The authors speculate that VLCFA incorporation into plasma membranes may contribute to destabilization and changes in membrane properties as reported [62], [63], [64] altering membrane microdomains (lipid rafts) [60] which would result in cell signaling disturbances, thus interfering with the turnover of NADPU p91PHOX [24].

Another potential source of ROS is the 5-lipoxygenase enzyme (5-LOX), which belongs to the arachidonic acid metabolizing family of inflammatory enzymes. This has been found elevated in cALD brain, even in intact areas, and its levels are increased depending on exogenous hexacosanoic acid levels [65]. 5-LOX converts products of arachidonic acid metabolism into leukotriene LT4. Leukotrienes are signaling molecules playing roles as chemoattractants or inducing cell death. Because leukotrienes modify cysteine residues of proteins/peptides, including glutathione (GSH), they are implicated in oxidative stress and subsequent inflammation [66]. Taken together, the available evidence discussed so far indicates that there are two sources of oxidative damage in X-ALD: i) inflammation, which entails the production of vast amounts of RNS and ROS from glial cells, NOS2, NADPH oxidase and 5-LOX being the primary sources, respectively; and ii) the chained dysfunction of peroxisomes and mitochondria leading to a severe rupture of redox homeostasis.

6. Inflammation and oxidative stress are intertwined

The observation that C26:0-accumulating lymphoblasts from X-ALD patients release pro-inflammatory cytokines along with ROS [24] points to a joint regulation of inflammation and oxidative stress by VLCFA. This is hardly surprising, since it is well established that inflammatory processes are tightly controlled by oxidative stress through the presence of redox sensors in key regulators of inflammatory responses (reviewed in [67]). One is the transcription factor NFκB, which has redox-reactive cysteine residues, the same as the I-κB kinase necessary for NFκB activation. These residues are susceptible to modification by reactive carbonyl compounds (e.g. hydroxynonenal, HNE), prostaglandins, NO and ROS. As advanced above, nitrosylation suppresses the translocation and DNA-binding of NFκB [68], while H2O2 activates these parameters in synergy with cytokines [69]. The duration of these interactions is exquisitely attuned to the redox potential established by the major redox-controlling systems, GSH/GSSG or thioredoxin [70], [71], [72], [73].

Another molecule characteristically regulated by oxidative stress is Nrf2, a transcription factor that coordinates the synthesis of antioxidant systems, reducing systems, and Nrf2 itself, via the antioxidant response element (ARE) present in the 5′ regulatory region of target genes [74]. Nrf2 is activated when oxidative stress modifies thiol groups in Keap1, thereby releasing the transcription factor. Neurons lacking Nrf2 are more susceptible to excitotoxic damage [75], indicating a key protective role of the factor under oxidative stress.

That two transcription factors with opposing roles can be regulated by oxidative stress unravels the intricate regulation of redox homeostasis. In this context, it is plausible to propose that high levels of ROS released by neurons, oligodendrocytes or other glial cells may activate an initially protective inflammatory response in glial cells via NFκB. Persistence or increased dose of the primary stimuli (e.g. VLCFA or cellular debris), or defective counteractive mechanisms (e.g. anti-inflammatory cytokines, or Nrf2-dependent protection) may push glia inflammation across an irreversible threshold, leading to aberrant activation as in cALD. The discovery of what drives the inflammatory reaction in AMN beyond homeostatic control to cause cAMN will obviously have important therapeutic implications.

7. Oxidative stress causes energy failure

Energy homeostasis and antioxidant defense cannot be considered independently as both are regulated by the ratios of redox-active cofactors NADH/NAD+, NADPH/NADP+ and GSH/GSSG [76]. NADH/NAD is linked to the synthesis of ATP via glycolysis and oxidative phosphorylation. NADH is produced by the TCA in mitochondria and is used as a substrate for complex I of the respiratory chain. The cellular ratio of NADPH/NADP+ in turn depends on the consumption of NADPH in antioxidative and biosynthetic enzyme reactions, and on the regeneration of NADPH by NADPH/regenerating enzymes. The GSH/GSSG ratio is determined by the NADPH consuming enzyme glutathione reductase, while glycolysis and the pentose phosphate cycle are predominantly responsible for the cytosolic reduction of NAD+ and NADP+, respectively.

In spinal cords from Abcd1 null mice we found diminished levels of NADH and ATP [32], most likely due to the aforementioned malfunction of TCA cycle and glycolitic enzymes resulting, in turn, from the early, pre-symptomatic oxidative damage to key proteins, thus confirming an overall bioenergetic failure (Fig. 1). Indeed, it has been suggested that oxidative modification of key mitochondrial TCA enzymes such as pyruvate and α-ketoglutarate dehydrogenases and aconitase may be an important pathophysiological factor in various neurodegenerative diseases [77], [78], [79]. GSH was diminished as expected under conditions prone to oxidative damage.

Oxidative phosphorylation (OXPHOS) in mitochondria is the main route to producing ATP. However, there was no impairment of OXPHOS activities in whole spinal cords [32], in agreement with previous results in isolated mitochondria from muscle of Abcd1 null and control mice [80]. These results suggest that metabolic failure is most likely due to impairment in glycolysis, although we cannot rule out the possibility that OXPHOS impairment in selected brain cells (e.g. neurons, see below) goes unnoticed in measurements in whole tissue. Spinal cords contain a mixture of gray and white matter. Neurons represent around 10% of the total amount of cells, whereas glia accounts for rest, with astrocytes being the most abundant cell type. Energy is mainly produced by mitochondria in neurons and by glycolysis in astrocytes in culture conditions [81]. Astrocytes export lactate (derived from glucose or glycogen) to neurons to power their mitochondria and support axonal function under conditions of energy deprivation [82]. Thus, a reduction of ATP in astrocytes would be owing to defective glycolysis, while in neurons it could be due to a reduction of NADH generation in mitochondria caused by the damage to KGDHC and aconitase and/or to some other unidentified enzymes. In a different neuropathological scenario, according to findings in a mouse model of Alzheimer’s disease (AD), glycolysis induction could be a mechanism to compensate for mitochondria dysfunction, in metabolic reprogramming [83]. This reprogramming would not be efficient in Abcd1 null spinal cords, as ALDO A, PFK1 and PKM2 are oxidized and their activity might be altered.

Energy deficiency has fatal consequences for axons because of their unusual size and high metabolic demands. It has been shown that reduced ATP production in affected neurons reduces their capacity to respond to the high energy demands of axonal transport and synaptic input. This may lead to axonal degeneration in our particular scenario, and also to neuronal demise in the most common neurodegenerative diseases [84].

8. Cascade of events leading to X-ALD: a model of disease pathogenesis

In Fig. 2 we present possible scenarios of cALD and AMN pathogenesis. We posit that inflammation and energy failure are the major players in X-ALD. In c-ALD, both inflammation and energy failure would overlap to cause oligodendrocyte and axonal death. The robust inflammatory reaction involves the release of toxic ROS and RNS as well deleterious cytokines like TNFα. The striking recovery of cALD patients observed after allogeneic hematopoietic stem cell transplantation, which gives rise to healthy microglia in the brain [85], does demonstrate a key role of microglia in the cerebral forms of X-ALD, possibly through abnormal cytokine and chemokine release of ABCD1-deficient microglial cells. An interesting twist on the regulation of inflammation in X-ALD is that VLCFA and other fatty acids (FA) may be signaling molecules connecting the immune system and the metabolism since pathways regulating metabolic and immune functions overlap [86]. This may allow nutrients such as fatty acids to act through pathogen-sensing systems such as Toll-like receptors (TLRs), giving rise to metabolic or nutritionally induced responses [87]. It is tempting to speculate that the fatty acids accumulated due to Abcd1 inactivation could activate TLR signaling, leading to NFκB-induced pro-inflammatory response in X-ALD, with the subsequent activation of pro-inflammatory cytokines.

By contrast, AMN has been traditionally considered a non-inflammatory condition. However, recent observations from our laboratory point to the existence of an early, pre-symptomatic and pro-inflammatory reaction in spinal cords of Abcd1 null mice, the model of adult AMN [21]. Inflammation hence does occur in supposedly non-inflammatory X-ALD phenotypes, although its protective or deleterious role remains unknown. Until this is clarified, axon degeneration can be attributed primarily to the energy failure, perhaps originating in the wrapping oligodendrocytes, in view of the finding that peroxisomes in oligodendrocytes play key roles in the maintenance of axons and myelin throughout adult life [56].

A note on the non-specific vs. receptor-mediated actions of VLCFA. The enrichment of membranes with VLCFA is the earliest biochemical abnormality in brain tissues from X-ALD patients [51], suggesting that this is the trigger for the ensuing pathogenic cascade. As noted, VLCFA are usual constituents of complex lipids such as gangliosides, proteolipids as well as the phosphatidylcholine and cholesterol esters that are characteristically abundant in myelin sheets. Biophysical approaches have revealed that VLCFA disrupts cellular membranes in a long-lasting manner, because the desorption rate of these molecules is slower than that of short chain fatty acids [63]. The incorporation of VLCFA to complex lipids might destabilize cell membranes resulting in cell signaling disturbances [62], [63], [64]. VLCFA may thus cause membrane destabilization of myelin leading to demyelination. Axonal degeneration can also be a consequence, because the integrity of the cellular membrane is a prerequisite for myelin–axonal interaction. However, at present, it is not known whether and how membrane disruption by VLCFA causes peroxisomal and mitochondrial dysfunction, oxidative stress, energy failure or inflammation, or acts as an additional etiopathogenic factor. The very early events in VLCFA-inflicted damage and X-ALD pathogenesis are particularly obscure.

In addition, research from our laboratories and other groups has hitherto provided a general picture of X-ALD pathogenesis with scant detail about cell compartmentalization. Above, we pointed out possible distributions of inflammatory or metabolic pathways among astrocytes and oligodendrocytes, but understanding of cell-specific actions of VLCFA, and the pathological engagements of neural cells, remain pending issues in X-ALD research that need to be addressed.

9. A hope for a treatment: antioxidants against axonal degeneration

Above, we outlined current knowledge on pathways leading to X-ALD (Fig. 1, Fig. 2). As noted, genetically-modified stem cells have yielded excellent results in cALD [85]. However, there is as yet no therapy to halt the slow progression of axonopathy in AMN, or to prevent the conversion of AMN into the deadly cAMN.

The very upstream position of oxidative stress in the pathogenic cascades brings this process to the center stage of AMN therapeutics. It is not trivial, however, how to tackle oxidative stress, since it encompasses a network of multiple signaling pathways. Ideally, reliable interactome maps should be used to select strategic network nodes leading to reduced ROS production, increased anti-oxidant protection, and improved mitochondrial function, leading to reversal of energetic failure. In the absence of cell-specific interactomes to model drug actions in AMN, a rougher but valid approach is to combine antioxidant therapies to target several spots with synergistic effects. In this line, we have recently undertaken a “proof-of-concept” study on the causative role of oxidative stress in axonopathy by testing a combination of anti-oxidants in Abcd1 null mice [88].

The cocktail of antioxidants contained: i) α-tocopherol, as it can inhibit the propagation phase of the peroxidative process by neutralizing the lipid-derived radicals [89]; ii) N-acetyl-cysteine, as it can regenerate reduced glutathione and scavenge several types of ROS including OH, H2O2, peroxyl radicals and nitrogen-centered free radicals [90]; and iii) α-lipoic acid, which can regenerate GSH from its oxidized counterpart, ascorbate from dehydroascorbate, and α-tocopherol from tocopherol radicals, thus enhancing the effects of the other two compounds [91]. In addition, α-lipoic acid and its reduced form, dihydrolipoic acid, may use their chemical properties as a redox couple to alter protein conformations by forming mixed sulfides, thus protecting proteins from oxidation. Further, since α-lipoic acid is an essential cofactor of PDHC and KGDHC, it could protect and increase the enzymatic activity of KGDHC [92], which suffers oxidative modifications in Abcd1 null mice, and therefore help bring about an increase in NADH and ATP production.

The three antioxidants were initially tested in X-ALD fibroblasts, where we observed a complete scavenging of VLCFA-dependent ROS generation when the drugs were used individually [88]. Importantly, the anti-oxidants displayed synergistic actions, which allowed using smaller doses of each oxidant in the combination than when tested individually. Use of several drugs with synergistic effects acting on unrelated targets may thus permit a considerable reduction in the dose of each individual compound, increasing selectivity and reducing adverse events. In addition, the combined anti-oxidant therapy is aimed at reproducing the multistep, combined response that is observed in vivo leading to recovery after an oxidative challenge [93]. Indeed, combinations of antioxidants have shown beneficial effects in pathologies associated with increased oxidative stress in mice [94] and in mitochondriopathies in human patients [95], [96].

The antioxidants were administered for 6 months to Abcd1 null mice between 12 and 18 months of age, that is, when oxidative damage is already taking place and histological and behavioral damage are on the verge of becoming evident. The results were spectacular, as the cocktail halted the histopathological signs of axonal degeneration and the onset of locomotor deficits [88]. Signs of oxidative damage to proteins were also abrogated, indicating that the antioxidants had truly engaged their targets. Indeed, oxidative damage to the five proteins differentially oxidized was erased. Metabolic failure reflected by the levels of ATP and ratios of NAD+/NADH was corrected, and the levels of pyruvate kinase and its enzymatic activity were restored, together with cellular contents of GSH [32]. Experiments were repeated in a double mutant lacking both Abcd1 and Abcd2 genes (Abcd1/abcd2 null), which develops an earlier and more severe axonal degeneration phenotype, thus being a more convenient model to assay therapeutic approaches [30]. Abcd2 and Abcd1 genes share overlapping functions regarding import of hexacosanoic and C26:1 fatty acids [97], [98], and remarkably, loss of Abcd2 causes oxidative damage in adrenal glands [99], while the loss of both transporters worsens oxidative damage and VLCFA accumulation [30]. The double mutant Abcd1/abcd2 null mice were treated after disease onset, at 12 months of age, during six months. The antioxidant treatment halted and reversed immunohistochemical signs of axonal degeneration and locomotor disability [88].

To our knowledge, this is the first demonstration that antioxidant therapy prevents or rescues axonal degeneration, and the results paved the way for a clinical trial in AMN patients that is now in progress.

Other combinations of antioxidants are most likely possible, including other compounds that target mitochondria, stimulate anti-oxidant defense and mitochondrion biogenesis at the level of transcription of relevant genes, e.g. via Nrf2/ARE, or increase ATP by mitochondria-independent sources. Clearly, more research is warranted to identify sources of ROS within the cell, and the exact mechanism of ROS production, in order to choose the best suited antioxidants from a growing panoply of compounds able to cross the blood–brain barrier. Because many antioxidants target mitochondria, it is of paramount importance to ascertain or to rule out the contribution of this organelle to ROS production in X-ALD. And within the mitochondria, it is important to identify the molecular source of ROS; i.e. whether the respiratory chain, TCA cycle enzymes and monoaminooxidases are relevant ROS sources.

In recent years, great attention has been focused on CoQ10, which participates in the electron transfer chain. CoQ10 may have, however, two major limitations as a therapy. First, the functioning of CoQ10 is dependent entirely on the functioning of the electron transfer chain which facilitates the redox cycling of the enzyme from ubiquinone (oxidized form) to ubiquinol (reduced form). This requirement becomes critical when oxidatively damaged mitochondria contain damaged electron transfer chain and thus cannot recycle CoQ10 appropriately. Second, there is evidence indicating that CoQ10 is unable to cross the blood–brain barrier [100], [101]. An alternative is the CoQ10 derivative MitoQ (Antipodean Pharmaceuticals Inc), a triphenylphosphonium-linked ubiquinone derivative [102]. MitoQ concentrates highly in the mitochondria because of the large mitochondrial membrane potential (Δψm) [103], and is also an effective antioxidant in the absence of a functioning electron transfer chain [104], [105].

10. Other neurodegenerative diseases as metabolic/inflammatory syndromes caused by oxidative damage

Chronic impairment of bioenergetics and mitochondria metabolism, together with oxidative stress, is a common noxa underlying age-related, multifactorial neurodegenerative diseases, such as PD, HD, and AD, to cite a few [16], [106], [107], [108], [109], [110], [111]. It has been postulated that oxidative stress in mitochondria can reduce the activities of various proteins due to oxidative modifications [77], [78], [79], [112], [113]. Both mtDNA and protein modifications have been described as resulting in a metabolic failure characterized by an increase in NAD+/NADH ratio (i.e., a decrease in cellular reducing potential), which is a powerful regulator of glycolysis, TCA cycle and oxidative phosphorylation [114], and also in reduced levels of ATP [84], [115]. Diminished levels of ATP have been described in AD and PD mouse models [116], [117], although the ratios of NAD/NADH have not been measured in most prominent neurodegenerative diseases. Most likely linked to the intertwined oxidative stress and metabolic derangement, such disorders also display early axonal pathologies including abnormal accumulation of proteins and organelles, and disrupted axonal transport, which is responsible for a substantial part of the debilitating symptoms that patients endure [118], see Fig. 3. Thus, therapeutic strategies that are of benefit for the above-mentioned diseases may deserve testing on X-ALD and vice versa.

10.1. Alzheimer’s disease

AD is the most common type of dementia, accounting for an estimated 60–80% of cases. Currently, over 30 million people worldwide have AD, and with the current steady increase in life-expectancy, the number of people with AD is forecast to increase to 1 out of 85 people by 2050. Memory loss is the most prevalent and early symptom, often accompanied by apathy and depression. Histological hallmarks in the brain are plaques of amyloid β (Aβ), and tangles of hyperphosphorylated tau. Axonal transport defects have been described as an early pathological feature in a variety of animal models [119], [120]. Current therapies are limited to drugs that attenuate disease symptomatology without addressing the causes of the disease.

It is widely considered that Aβ and tau hyperphosphorylation, which may be causally related, are culprits in the disease, with inflammation, oxidative stress and cerebrovascular factors acting as amplifying elements. However, there has been a failure of over twenty clinical trials aimed at testing drugs against alleged causes of the disease. Failed drugs include anti-amyloid therapies like the γ-secretase inhibitor semagacestat and the γ-secretase modulator tarenflurbil, non-steroidal anti-inflammatory drugs (NSAIDs), estrogen, valproate, the insulin-sensitizers rosiglitazone and pioglitazone, the cholesterol-reducing drug simvastatin, the antioxidants omega 3/docosahexaenoic acid (DHA), Gingko biloba, curcumin, and vitamin E (clinicaltrials.gov). While careful appraisal of each trial reveals on occasion lack of sound animal data, or evidence that the drug engaged its target in the brain, the prevailing view is that treatments were tested too late, when damage is perhaps irreversible. The recognition that AD progresses in a clinically silent manner over years –perhaps two decades– before clinical symptoms are manifested is arguably the most important recent breakthrough in AD [121], [122]. This vision implies that treatments will have to strictly match the onset and temporal evolution of their targets. “Good timing” in drug administration emerges as a key strategic factor in AD therapeutics. The problem arises that the chronological appearance and interplay of pathogenic cascades during pre-clinical stages are not known. The recent guidelines issued by the National Institute for Aging (NIA) and the Alzheimer Association (AA) in the USA provide a valuable framework for exploring pre-clinical stages [122]. Abnormal brain deposition of Aβ detected with PET or measurements in CSF, appears to be the earliest pathological hallmark, followed by neurodegeneration and, later, a very subtle cognitive decline before entering the clinical categories of early AD (e.g. Subjective Clinical Complaint (SCI) and Mild Cognitive Impairment (MCI)) [122]. Where in this picture oxidative stress lies, and how it interacts with Aβ tau or inflammation, are outstanding questions in AD.

The presence of oxidative stress and oxidative modifications in AD has been overwhelmingly recognized for years, and it has been the topic of a wealth of studies and reviews [111]. In brief, oxidative modifications have been detected in postmortem samples from AD patients affecting DNA [123], along with proteins implicated in glycolysis and energy metabolism, mitochondrial electron transport chain and oxidative phosphorylation, structural proteins, chaperones, stress proteins, and ubiquitin-proteasome system components (see Table 1) [111], although loss-of-function analysis of oxidized proteins in AD is for the most part pending. The fact that dozens of proteins have been reported oxidized in AD, but only five in X-ALD [32], may have several explanations. One is the differing stages in disease progression of the samples used: most of the postmortem samples of AD were obtained from patients, that is, from late disease, while the analysis in X-ALD was performed at pre-symptomatic stages in Abcd1 null mice. Note also that different species were used in the analyses. Another explanation is that there may be a disease-specific oxidative damage that primarily affects energy metabolism in X-ALD, while in AD it is more widespread; however, ATP synthase, the enzyme that catalyzes the synthesis of ATP from ATP and inorganic phosphate, is consistently found oxidized in AD [124], [125], [126], but was not detected with redox proteomics in X-ALD mice, pointing to further unclear differences in the compartmentalization of pathogenic pathways between AD and X-ALD.

Table 1. Targets of oxidative damage in adrenoleukodystrophy (X-ALD), Alzheimer’s disease (AD), Huntington’s disease (HD) and Parkinson’s disease (PD). Proteins found oxidized at least in two of the diseases are shown.

Short nameFunctionX-ALDADHDPD
Pyruvate kinase isozymes M1/M2KPYMGlycolysisXX
Phosphoglycerate kinase 1PGK1GlycolysisXXX
Fructose-biphosphate aldolase AALDOAGlycolysisXXXX
Glyceraldehyde-3-phosphate dehydrogenaseG3P, GAPDHGlycolysisXX
Lactate dehydrogenaseLDHGlycolysisXX
Aconitase hydratase, mitochondrialAconitaseKrebs cycleXXXX
Dihydrolipoyl dehydrogenaseDLDHKrebs cycleXX
Creatine kinaseCKEnergy transductionXX
Outer mitochondrial membrane protein porin 1VDACIon transporterXX
Superoxide dismutase 1SOD1Antioxidant responseXX
Parkinson disease protein 7DJ-1Antioxidant responseXX
Peroxiredoxin 2NKEF-BAntioxidant responseXX
Tubulin beta-2A chainTBB2ACytoskeletonXX
Glial fibrillary acidic proteinGFAPCytoskeletonXX
Carbonic anhydrase 2CAH2, CAIICO2 metabolismXX

Mitochondrial dysfunction and morphological alterations are also hallmarks of AD [127], [128]. Several antioxidant drugs targeting mitochondria are under investigation for the treatment of AD [129]. CoQ10 has been shown to be protective in transgenic models of AD [130], [131], but we saw above that the problem of crossing the blood–brain barrier impedes the use of this drug to treat brain diseases [100], [101]. N-acetyl-cysteine, acetyl-l-carnitine (ALCAR), r-α-lipoic, polyphenols and tert-butylhydroquinone (TBHQ), alone or in combination, are protective in cellular and transgenic models of AD, and in aged animals [132], [133]. It has to be stressed that experimental paradigms in animals are usually preventive, since drugs are administered before age- or AD-related damage occurs. Thus, these anti-oxidant agents hold promise provided that they are administered preventively to AD patients. Of note, in pilot studies with twelve AD patients, efficacy of a nutraceutical formulation containing N-acetyl-cysteine, vitamin E and α-lipoic has been reported [134], [135]. Larger trials are clearly necessary.

The interaction of oxidative stress and inflammation in AD is not well understood. The existence of multiple inflammatory mediators has been widely documented in animal models and in body fluids and postmortem brains from AD patients [136]. The collected data have propelled a view in which very toxic byproducts of microglia activation, namely free radicals, are a major cause of AD as in the cerebral forms of X-ALD. However, there are major caveats attaching to this conclusion (reviewed in [137]). There is a need for better cellular and animal models recapitulating inflammation in aged brains, particularly in the preclinical stages, before any immunomodulatory therapy is designed for AD. This situation mirrors the one existing in X-ALD, where an animal model of inflammatory demyelination as it occurs in the child variants of disease is still lacking. However, as described in Section 8, our recent studies have unraveled the existence of a pre-symptomatic pro-inflammatory cascade in brains from cALD and cAMN patients, which is reproduced by the mouse model [21]. At present, the pathological role of this phenomenon is not clear. A key question is if a basal inflammatory reaction may facilitate the outburst of a lethal demyelinating inflammation, particularly in adults. Conveniently manipulated, the mouse model may be of help to answer these questions.

By contrast, a large body of evidence in cells and animal models supports the idea that Aβ and oxidative stress are highly interactive. Aβ binds to mitochondrial membranes and interferes with the normal electron flow through the respiratory chain; this results in defective mitochondrial energy metabolism and increased ROS production [138], [139], [140], reviewed in [141]. Conversely, oxidative stress drives Aβ production and impairs Aβ clearance in vitro and in vivo [142], [143], [144], [145]. Of note, there is an active γ secretase in mitochondria [146], reinforcing the view that Aβ production is closely linked to mitochondrial activity. Interestingly, a recently elaborated interactome of ApoE has revealed possible actions of the lipoprotein at the mitochondria [147], suggesting that ApoE4 may increase the risk of AD by altering mitochondrial functions.

All in all, these data suggest that abnormal Aβ accumulation and oxidative stress may be synchronous in AD progression. Since 20–40% of healthy individuals have plaques in the brain after the age of 65 [121], and projections from PET-based measurements of Aβ plaque load suggest that the abnormal deposition may start up to 20 years before the dementia onset [148], it may be reasoned that oxidative stress is one of the earliest events in AD. Indeed, ATP synthase was found oxidized and its activity reduced in asymptomatic cases with AD pathology in Braak II stage [149]. Hypometabolism, a plausible marker of mitochondrial dysfunction as revealed by PET measurement of fluorodeoxyglucose (FDG), is a very sensitive pre-symptomatic marker of AD-related neurodegeneration [150]. In AD-mice, lipid peroxidation and mitochondrial alterations precede Aβ plaque and tau deposition [151], [152]. This evidence supports a scenario in which the interplay between oxidative stress and Aβ oligomers/fibrils, the initial culprits of AD and upstream triggers of tau pathology, may further exacerbate Aβ and tau aggregation.

While more research is clearly necessary to characterize oxidative stress in AD, it appears reasonable that Aβ and oxidative stress, or mitochondrial damage, be therapeutically targeted in unison at pre-symptomatic stages. Anti-amyloid therapy, e.g. immunotherapy or Notch-sparing γ secretases, should be combined with a cocktail of antioxidants as in X-ALD. Candidates include antioxidants supported by the latest animal data (e.g. N-acetyl-cysteine and α-lipoic acid) and those failing in clinical trials (e.g. vitamin E), under the assumption that these treatments may not have worked because they were administered alone or too late. Finally, measurements in CSF of isoprostane, currently the most reliable marker of oxidative stress in fluids [153], may help to yield further insight into the temporal relationship between Aβ and oxidative stress, and to guide antioxidant therapies in AD.

10.2. Parkinson’s disease

PD is the second most common neurodegenerative disease, affecting over ten million people worldwide. Although PD is most common in people over 60, many people are diagnosed in their 40s and younger. PD is characterized by a marked loss of midbrain dopaminergic neurons localized in the substantia nigra pars compacta. This leads to progressive rigidity, bradykinesia and tremors. Abnormal aggregations of α-synuclein (i.e. Lewy bodies) are considered a major pathogenic factor in the disease, and indeed the most recent animal models encompass transgenic over-expression or viral delivery of α-synuclein. Autosomal dominant factors involved in early-onset familial PD include mutations in the SNCA (α-synuclein), and LRRK2 (leucine-rich repeat kinase 2) genes. Autosomal recessive factors include mutations in the PARK2 (parkin), PINK1 (PTEN induced putative kinase 1), PARK7 (DJ1), and ATP13A2 (ATPase type 13A2) genes. Gene wide genomic associations (GWAS) have revealed that SCNA and LRRK2, among other genes, confer susceptibility to late-onset, sporadic PD [154].

Models of PD, both sporadic and caused by 1-methyl-4-phenylpyridinium (MPP+) proceed through a “dying back” pattern of neuronal degeneration characterized by early loss of synaptic terminals and axonopathy, with impairment of fast axonal transport [155]. Numerous studies support the idea that oxidative stress and mitochondrial dysfunction underlie the development of neuropathology, and there are excellent reviews on the topic [111], [156], [157]. Herein we will pinpoint key distinctive facts pertaining to antioxidant treatment in the disease:i)

Mitochondrial dysfunction appears as an early, upstream if not causative factor in PD pathogenesis, as judged by the role of Parkinsonism-linked genes, which maintain mitochondrial integrity by regulating diverse aspects of mitochondrial function, including membrane potential, calcium homeostasis, cristae structure, respiratory activity, mtDNA integrity, and autophagy-dependent clearance of dysfunctional mitochondria. DJ1 appears to protect mitochondria against oxidative stress caused by sustained entrance of Ca2+ via L-type channels [158]. Loss of PINK1 causes defective oxidative phosphorylation, production of ROS and a decrease in mitochondrial content [117], [159]. Parkin, a molecule associated with the outer mitochondrial membrane, is an E3 ligase that targets proteins for degradation in the proteasome or lysosome [160]. Evidence indicates that parkin may be crucial for autophagy-dependent clearance of dysfunctional mitochondria [161]. LRRK2 interact with human peroxiredoxin 3, a mitochondrial member of the antioxidant family of thioredoxin peroxidases, and mutations of LRKK2 promote peroxidase dysfunction, dysregulation of mitochondrial function and oxidative stress [162]. Finally, α-synuclein, although predominantly cytosolic, is detected in mitochondria where it inhibits complex I, and reduces ATP synthesis and membrane potential [163]. α-Synuclein also inhibits mitochondrial fusion, and this effect is rescued by parkin, DJ1 and PINK1 [164], [165], further supporting the role of these proteins in regulating mitochondrial dynamics. Damage to mitochondria has been shown to perturb transport of mitochondria through axons [166].ii)

Complex I deficiency and glutathione depletion have been found in the substantia nigra of sporadic and pre-symptomatic PD patients [167].iii)

Numerous studies have described oxidative modifications in PD and in animal models (reviewed in [111]). There is oxidative modification selective to mitochondria-associated metabolic proteins (LDH, enolasa and CA II anhydrase), concomitant to loss of function, in transgenic mice overexpressing the human A30P α-synuclein mutation [168], see Table 1. In patient brains, in addition to oxidative modification and functional impairment of complex I [169], there are two remarkable findings. One is that proteins that confer causality or susceptibility to familial or sporadic PD are targets of oxidative stress. This is the case of α-synuclein [170], parkin [171], and DJ1 [172]. A vicious circle may exist whereby dysfunction of these proteins causes mitochondrial damage and a ROS increase, which damages PD-related proteins further. The second remarkable finding is that oxidative modifications have been detected not only in the substantia nigra [170], but also in the cerebral cortex and the amygdala [170], [173], at pre-clinical pre-motor stages of PD (stages II and III in Braak classification). This reinforces the emerging view that PD, like AD, develops for many years without marked symptoms and that, aside from the midbrain, the disease affects other brain areas not associated with Lewy bodies [173], [174], [175].iv)

A recent study has implicated down-regulation of two microRNAs, miR-34b and miR-34c, in several brain areas with variable neuropathological involvement at clinical (motor) stages (Braak stages IV and V) and pre-motor stages (stages I–III) of the disease. According to additional data, these microRNAs may be related to mitochondrial function and dynamics, as well as redox regulation, via DJ1 and parkin [176].v)

Auto-oxidation of dopamine is an important source of intracellular ROS production [177]. The direct product of such auto-oxidation reactions is the superoxide radical, which may contribute to the anatomically-specific degeneration of dopaminergic neurons in PD.

Restitution of dopamine or dopaminergic activity is the only available therapy for PD. In view of the aforementioned evidence, and in the absence of disease-modifying therapies targeted to Parkinsonism-linked proteins, antioxidant therapies aimed at restoring mitochondrial function and the reduced glutathione, or ROS scavengers, remain an option for combination with dopamine analogs. Indeed, intravenous N-acetyl-cysteine, intranasal glutathione, MitQ, and the PPAR gamma and Nrf2/ARE activator pioglitazone are being tested in PD patients at phase I or II (clinicaltrials.gov). A phase IV has been completed with levodope plus rasagiline, a monoaminooxidase B inhibitor with antioxidant properties, but the results have not been disclosed yet (clinicaltrials.gov). However, a recent phase III trial with CoQ10/vitamin E was terminated with the statement that “The investigational drug is unlikely to demonstrate efficacy over placebo for this indication” (clinicaltrials.gov). There are two, not incompatible explanations for this failure. One, only more complex agent combinations addressing in unison dopamine depletion and several targets of mitochondrial dysfunction, as well as mitochondria biogenesis, will show therapeutic benefits. Two, patients may have been treated too late; that is, disease-modifying interventions should be moved to the pre-motor stages of PD. This will require pre-clinical biomarkers, which are not currently available. In this regard, the recent detection of α-synuclein oligomers in the CSF of PD patients offers hope that the misfolded protein may become a clinical and pre-clinical marker for PD, as Aβ is for AD [178].

10.3. Huntington’s disease

HD is an autosomal dominantly inherited neurodegenerative disease caused by the expansion of CAG trinucleotide repeats in the huntingtin gene. This gives rise to an elongated polyglutamine sequence (polyQ) at the N-terminal of the huntingtin protein. The disease starts when the number of glutamines expands beyond 40, and it is characterized by atrophy of the striatum, intraneuronal huntingtin aggregates, reactive gliosis and, as clinical symptoms, involuntary movements (chorea), psychiatric disturbances or dementia, and death after 15–20 years of progression. While the function of normal huntingtin is not known, polyQ-huntingtin disrupts many cellular processes: autophagy, energy metabolism, gene transcription, clathrin-dependent endocytosis, intraneuronal trafficking, postsynaptic signaling and axonal transport [179]. We list below key evidence in support of the view that oxidative stress, mitochondrial dysfunction and impaired energy metabolism contribute to HD (reviewed in [180]):i)

PolyQ-huntingtin causes ROS production [181], and there are mitochondrial defects in HD patients [182]. Indeed, polyQ-huntingtin interacts with mitochondrial proteins, binding to the mitochondrial fission GTPase dynamin-related protein-1 (DRP1) in mice and humans with HD, which, in turn, stimulates its enzymatic activity leading to impairment of mitochondrial fission–fusion balance, mitochondria fragmentation and neuronal injury [128], [183]. In addition, mutant huntingtin represses the expression of PPARγ, coactivator 1 α (PGC-1α), a transcriptional coactivator that regulates several metabolic processes, including mitochondrial biogenesis and respiration [180]. Accordingly, PGC1-α knockouts exhibit impaired mitochondrial function, hyperkinetic movement and striatal degeneration, as seen in HD [191].ii)

Complex II activity is decreased in the HD brain, and complex II inhibition with 3-nitropropionic acid induces striatal degeneration and movement disorders in rodents and primates [184], that is, it induces a clinical phenotype very reminiscent of that in HD.iii)

Redox proteomics identified seven proteins in HD brains involved in glucose metabolism and mitochondrial energy pathways affected by oxidative modifications see Table 1 [185]. Interestingly, oxidation also affected the activity of enzymes involved in the metabolism of pyridoxal 5-phosphate (PLP), the active form of vitamin B6 [185]. PLP acts as a coenzyme in all transamination reactions and some decarboxylation and deamination reactions, including the conversion of L-DOPA into dopamine and glutamate into GABA. Thus, synthesis of dopamine and GABA may be impaired, and glutamate accumulates abnormally in HD. Also, there is reduction of antioxidant systems in brains from HD patients [186].

These data argue in favor of antioxidant therapies specifically targeting mitochondria. CoQ10 [187], l-carnitine [188], and the Nrf2/ARE activators triterpenoids [189] have shown protective effects in animal models of HD. There is currently a phase III trial with CoQ10 and a phase II with l-creatine (clinicatrials.gov). There is no combination of antioxidants being currently tested, unfortunately. Finally, the importance of early treatment applies to HD as well. The fact that the disease can be diagnosed by predictive genetic testing provides a window of opportunity for intervention aimed at preventing or delaying disease onset. Biomarkers are therefore needed to monitor treatment efficacy. Clinical, cognitive, neuroimaging, and biochemical biomarkers are thus being investigated for their potential in pre-clinical use for patients with Huntington’s disease [190].

11. Concluding remarks

Although the first hits in the most prevalent neurodegenerative conditions and X-ALD are of different etiology and target different cell types, oxidative stress and associated bioenergetic failure involving mitochondria dysfunction, together with axonal damage, is a common bottleneck that justifies the undertaking of similar therapeutic strategies. Identifying the molecular mechanisms of ROS generation and mitochondrial demise in each specific disease is necessary to assist in making the best choice among neuroprotective approaches. Regarding antioxidant treatment, strategies should basically follow two routes. One is translational, aimed at obtaining sound animal data supporting the efficacy of multiple-drug combinations, thereby facilitating clearance by government agencies. Disease-specific modifying therapies should be combined with anti-oxidant therapies, the latter containing agents targeting membranes, mitochondria function and biogenesis as well as ROS scavengers. Optimal combinations and doses have to be empirically determined. Because a wealth of evidence indicates that oxidative stress is a primary pathogenic factor in neurodegeneration, it is of utmost importance that drugs be tested as early as possible in the disease progression, which requires validation of surrogate disease biomarkers. In X-ALD, longitudinal assessment of oxidative lesions in blood cells, together with clinical symptoms, will serve to monitor efficacy in an ongoing clinical trial testing a cocktail of antioxidants in AMN patients. In AD, the recently reported success of a preclinical trial with donepezil in cognitively normal subjects paves the way for preclinical trials. However, in PD and HD, biomarkers are still needed to enroll asymptomatic individuals and/or to assess disease progression. The other strategic route should lead to further knowledge concerning on the underpinnings of oxidative stress in inflammation and neurodegeneration. An outstanding question points to disease specific pathways, organelles, cells and brain areas implicated. Despite the many common themes shared by the diseases, we still do not know why for example, complex I or complex II dysfunction leads, respectively, to PD or HD, at least in experimental models. Nor do we know exactly how VLCFA, Aβ, or mutant α-synuclein or huntingtin cause mitochondrial damage, and why the resulting pathologies are vastly different. This intriguing fact may plausibly be due to a yet unexamined mitochondrial and metabolic heterogeneity throughout the brain, which determines a region-specific vulnerability to noxious agents. The devil is in the details when tailoring effective disease-targeted therapies.


The work of the authors of this review was supported by grants from the European Commission (FP7-241622), the European Leukodystrophy Association (ELA2009-041D6; ELA2008-040C4), the Oliver’s Army, the Spanish Institute for Health Carlos III (FIS PI080991), and the Autonomous Government of Catalonia (2009SGR85) to A.P. Centro de Investigación en Red sobre Enfermedades Raras (CIBERER) is an initiative of the Instituto de Salud Carlos III. The study was developed under the COST action BM0604 (to A.P.). Work carried out at the Department of Experimental Medicine was supported in part by R+D grants from the Spanish Ministry of Science and Innovation [BFU2009-11879/BFI], the Spanish Ministry of Health [PI081843, PI0111532], the Autonomous Government of Catalonia [2009SGR735], ‘La Caixa’ Foundation, and COST B35 Action of the European Union to M.P.O. and R.P. We are indebted to T Yohannan for editorial assistance. The authors declare no conflict of interest.


[1]J. Mosser, A.M. Douar, C.O. Sarde, P. Kioschis, R. Feil, H. Moser, A.M. Poustka, J.L. Mandel, P. AubourgNature, 361 (1993), pp. 726-730View Record in Scopus[2]C.W. van Roermund, W.F. Visser, L. Ijlst, A. van Cruchten, M. Boek, W. Kulik, H.R. Waterham, R.J. WandersFASEB J., 22 (2008), pp. 4201-4208View Record in Scopus[3]I. Ferrer, P. Aubourg, A. PujolBrain Pathol, 20 (2010), pp. 817-830View Record in Scopus[4]H. Moser, K.D. Smith, P.A. Watkins, J. Powers, A.B. MoserC. Scriver (Ed.), The Metabolic and Molecular Bases of Inherited Disease, McGraw-Hill, New-York (2001), pp. 3257-3301Google Scholar[5]B.M. van Geel, L. Bezman, D.J. Loes, H.W. Moser, G.V. RaymondAnn. Neurol., 49 (2001), pp. 186-194View Record in Scopus[6]B. Halliwell and J.M. Gutteridge, (2007).Google Scholar[7]A. Naudi, M. Jove, D. Cacabelos, O. Ramirez, V. Ayala, E. Ilieva, A. Kichev, I. Ferrer, M. Portero-Otin, R. PamplonaLipid Peroxidation: Biological ImplicationsA. Catala (Ed.), Angel Catala (2011), pp. 47-70View Record in ScopusGoogle Scholar[8]R. Pamplona, D. CostantiniAm. J. Physiol. Regul. Integr. Comp. Physiol., 301 (2011), pp. 843-863[9]S.J. HealesLancet, 357 (2001), p. 314ArticleDownload PDFView Record in Scopus[10]H. Sies, W. Stahl, A.R. SundquistAnn. N. Y. Acad. Sci., 669 (1992), pp. 7-20CrossRefView Record in Scopus[11]V.J. Thannickal, B.L. FanburgAm. J. Physiol. Lung Cell. Mol. Physiol., 279 (2000), pp. L1005-L1028CrossRefView Record in Scopus[12]B. Daneshvar, H. Autrup, L.O. DragstedBiomarkers (1997), pp. 117-123CrossRefView Record in Scopus[13]D. Boyd-Kimball, A. Castegna, R. Sultana, H.F. Poon, R. Petroze, B.C. Lynn, J.B. Klein, D.A. ButterfieldBrain Res., 1044 (2005), pp. 206-215ArticleDownload PDFView Record in Scopus[14]R. PamplonaChem. Biol. Interact., 192 (2011), pp. 14-20ArticleDownload PDFView Record in Scopus[15]S.R. Thorpe, J.W. BaynesAmino Acids, 25 (2003), pp. 275-281View Record in Scopus[16]M.T. Lin, M.F. BealNature, 443 (2006), pp. 787-795CrossRefView Record in Scopus[17]W.A. Pryor, G.L. SquadritoAm. J. Physiol., 268 (1995), pp. L699-L722CrossRefView Record in Scopus[18]J.D. Hayes, J.U. Flanagan, I.R. JowseyAnnu. Rev. Pharmacol. Toxicol., 45 (2005), pp. 51-88CrossRefView Record in Scopus[19]A.G. Gilg, A.K. Singh, I. SinghJ. Neuropathol. Exp. Neurol., 59 (2000), pp. 1063-1069CrossRefView Record in Scopus[20]J.M. Powers, Z. Pei, A.K. Heinzer, R. Deering, A.B. Moser, H.W. Moser, P.A. Watkins, K.D. SmithJ. Neuropathol. Exp. Neurol., 64 (2005), pp. 1067-1079CrossRefView Record in Scopus[21]A. Schluter, L. Espinosa, S. Fourcade, J. Galino, E. Lopez, E. Ilieva, L. Morato, M. Asheuer, T. Cook, A. McLaren, J. Reid, F. Kelly, S. Bates, P. Aubourg, E. Galea, A. PujolHum. Mol. Genet., 21 (2012), pp. 1062-1077CrossRefView Record in Scopus[22]M. Deon, A. Sitta, A.G. Barschak, D.M. Coelho, M. Pigatto, G.O. Schmitt, L.B. Jardim, R. Giugliani, M. Wajner, C.R. VargasInt. J. Dev. Neurosci., 25 (2007), pp. 441-444ArticleDownload PDFCrossRefView Record in Scopus[23]C.R. Vargas, M. Wajner, L.R. Sirtori, L. Goulart, M. Chiochetta, D. Coelho, A. Latini, S. Llesuy, A. Bello-Klein, R. Giugliani, M. Deon, C.F. MelloBiochim. Biophys. Acta, 1688 (2004), pp. 26-32ArticleDownload PDFView Record in Scopus[24]T. Uto, M.A. Contreras, A.G. Gilg, I. SinghDev. Neurosci., 30 (2008), pp. 410-418CrossRefView Record in Scopus[25]S. Fourcade, J. Lopez-Erauskin, J. Galino, C. Duval, A. Naudi, M. Jove, S. Kemp, F. Villarroya, I. Ferrer, R. Pamplona, M. Portero-Otin, A. PujolHum. Mol. Genet., 17 (2008), pp. 1762-1773CrossRefView Record in Scopus[26]J.F. Lu, A.M. Lawler, P.A. Watkins, J.M. Powers, A.B. Moser, H.W. Moser, K.D. SmithProc. Natl. Acad. Sci. U. S. A., 94 (1997), pp. 9366-9371View Record in Scopus[27]S. Forss-Petter, H. Werner, J. Berger, H. Lassmann, B. Molzer, M.H. Schwab, H. Bernheimer, F. Zimmermann, K.A. NaveJ. Neurosci. Res., 50 (1997), pp. 829-843View Record in Scopus[28]T. Kobayashi, N. Shinnoh, A. Kondo, T. YamadaBiochem. Biophys. Res. Commun., 232 (1997), pp. 631-636ArticleDownload PDFView Record in Scopus[29]A. Pujol, C. Hindelang, N. Callizot, U. Bartsch, M. Schachner, J.L. MandelHum. Mol. Genet., 11 (2002), pp. 499-505View Record in Scopus[30]A. Pujol, I. Ferrer, C. Camps, E. Metzger, C. Hindelang, N. Callizot, M. Ruiz, T. Pampols, M. Giros, J.L. MandelHum. Mol. Genet., 13 (2004), pp. 2997-3006View Record in Scopus[31]R. Sultana, D.A. ButterfieldFree Radic. Biol. Med., 50 (2011), pp. 487-494ArticleDownload PDFView Record in Scopus[32]J. Galino, M. Ruiz, S. Fourcade, A. Schluter, J. Lopez-Erauskin, C. Guilera, M. Jove, A. Naudi, E. Garcia-Arumi, A.L. Andreu, A.A. Starkov, R. Pamplona, I. Ferrer, M. Portero-Otin, A. PujolAntioxid. Redox Signal., 15 (2011), pp. 2095-2107CrossRefView Record in Scopus[33]S. Fourcade, M. Ruiz, C. Guilera, E. Hahnen, L. Brichta, A. Naudi, M. Portero-Otin, G. Dacremont, N. Cartier, R. Wanders, S. Kemp, J.L. Mandel, B. Wirth, R. Pamplona, P. Aubourg, A. PujolHum. Mol. Genet., 19 (2010), pp. 2005-2014CrossRefView Record in Scopus[34]A. Boveris, N. Oshino, B. ChanceBiochem. J., 128 (1972), pp. 617-630CrossRefView Record in Scopus[35]N.E. Tolbert, E. EssnerJ. Cell Biol., 91 (1981), pp. 271s-283sCrossRef[36]H. Singh, N. Derwas, A. PoulosArch. Biochem. Biophys., 259 (1987), pp. 382-390ArticleDownload PDFView Record in Scopus[37]E. Braschi, V. Goyon, R. Zunino, A. Mohanty, L. Xu, H.M. McBrideCurr. Biol., 20 (2010), pp. 1310-1315ArticleDownload PDFView Record in Scopus[38]J.D. LinMol. Endocrinol., 23 (2009), pp. 2-10CrossRefView Record in Scopus[39]H.R. Waterham, J. Koster, C.W. van Roermund, P.A. Mooyer, R.J. Wanders, J.V. LeonardN. Engl. J. Med., 356 (2007), pp. 1736-1741View Record in Scopus[40]F. Camoes, N.A. Bonekamp, H.K. Delille, M. SchraderJ. Inherit. Metab. Dis., 32 (2009), pp. 163-180CrossRefView Record in Scopus[41]J.I. Koepke, C.S. Wood, L.J. Terlecky, P.A. Walton, S.R. TerleckyToxicol. Appl. Pharmacol., 232 (2008), pp. 99-108ArticleDownload PDFView Record in Scopus[42]O. Ivashchenko, P.P. Van Veldhoven, C. Brees, Y.S. Ho, S.R. Terlecky, M. FransenMol. Biol. Cell., 22 (2011), pp. 1440-1451View Record in Scopus[43]R. Dirkx, I. Vanhorebeek, K. Martens, A. Schad, M. Grabenbauer, D. Fahimi, P. Declercq, P.P. Van Veldhoven, M. BaesHepatology, 41 (2005), pp. 868-878CrossRefView Record in Scopus[44]S. Goldfischer, C.L. Moore, A.B. Johnson, A.J. Spiro, M.P. Valsamis, H.K. Wisniewski, R.H. Ritch, W.T. Norton, I. Rapin, L.M. GartnerScience, 182 (1973), pp. 62-64View Record in Scopus[45]I. Ferrer, J.P. Kapfhammer, C. Hindelang, S. Kemp, N. Troffer-Charlier, V. Broccoli, N. Callyzot, P. Mooyer, J. Selhorst, P. Vreken, R.J. Wanders, J.L. Mandel, A. PujolHum. Mol. Genet., 14 (2005), pp. 3565-3577CrossRefView Record in Scopus[46]T. Cocco, M. Di Paola, S. Papa, M. LorussoFree Radic. Biol. Med., 27 (1999), pp. 51-59ArticleDownload PDFView Record in Scopus[47]P. Schonfeld, G. ReiserJ. Biol. Chem., 281 (2006), pp. 7136-7142CrossRefView Record in Scopus[48]S. Hein, P. Schonfeld, S. Kahlert, G. ReiserHum. Mol. Genet., 17 (2008), pp. 1750-1761CrossRefView Record in Scopus[49]M. Garashi, D. Belchis, K. SuzukiJ. Neurochem., 27 (1976), pp. 327-328[50]W.C. Hubbard, A.B. Moser, A.C. Liu, R.O. Jones, S.J. Steinberg, F. Lorey, S.R. Panny, R.F. Vogt Jr., D. Macaya, C.T. Turgeon, S. Tortorelli, G.V. RaymondMol. Genet. Metab., 97 (2009), pp. 212-220ArticleDownload PDFView Record in Scopus[51]C. Theda, A.B. Moser, J.M. Powers, H.W. MoserJ. Neurol. Sci., 110 (1992), pp. 195-204ArticleDownload PDFView Record in Scopus[52]M. Khan, J. Singh, I. SinghJ. Neurochem., 106 (2008), pp. 1766-1779View Record in Scopus[53]P. Brites, P.A. Mooyer, L. El Mrabet, H.R. Waterham, R.J. WandersBrain, 132 (2009), pp. 482-492View Record in Scopus[54]H. Hayashi, M. OohashiBiochim. Biophys. Acta, 1254 (1995), pp. 319-325ArticleDownload PDFView Record in Scopus[55]O. Krysko, A. Bottelbergs, P. Van Veldhoven, M. BaesMol. Genet. Metab., 100 (2010), pp. 71-76ArticleDownload PDFView Record in Scopus[56]C.M. Kassmann, C. Lappe-Siefke, M. Baes, B. Brugger, A. Mildner, H.B. Werner, O. Natt, T. Michaelis, M. Prinz, J. Frahm, K.A. NaveNat. Genet., 39 (2007), pp. 969-976CrossRefView Record in Scopus[57]I. Singh, A. PujolBrain Pathol, 20 (2010), pp. 838-844View Record in Scopus[58]A.Y. Abramov, J. Jacobson, F. Wientjes, J. Hothersall, L. Canevari, M.R. DuchenJ. Neurosci., 25 (2005), pp. 9176-9184CrossRefView Record in Scopus[59]R. DringenAntioxid. Redox Signal., 7 (2005), pp. 1223-1233CrossRefView Record in Scopus[60]D. Shao, A.W. Segal, L.V. DekkerFEBS Lett., 550 (2003), pp. 101-106ArticleDownload PDFView Record in Scopus[61]X.L. Cui, J.G. DouglasProc. Natl. Acad. Sci. U. S. A., 94 (1997), pp. 3771-3776View Record in Scopus[62]R.A. Knazek, W.B. Rizzo, J.D. Schulman, J.R. DaveJ. Clin. Invest., 72 (1983), pp. 245-248CrossRefView Record in Scopus[63]J.K. Ho, H. Moser, Y. Kishimoto, J.A. HamiltonJ. Clin. Invest., 96 (1995), pp. 1455-1463CrossRefView Record in Scopus[64]R.W. Whitcomb, W.M. Linehan, R.A. KnazekJ. Clin. Invest., 81 (1988), pp. 185-188CrossRefView Record in Scopus[65]M. Khan, J. Singh, A.G. Gilg, T. Uto, I. SinghJ. Lipid Res., 51 (2010), pp. 1685-1695CrossRefView Record in Scopus[66]X.J. Huang, W.P. Zhang, C.T. Li, W.Z. Shi, S.H. Fang, Y.B. Lu, Z. Chen, E.Q. WeiGlia, 56 (2008), pp. 27-37CrossRefView Record in Scopus[67]S. Rosales-Corral, R.J. Reiter, D.X. Tan, G.G. Ortiz, G. Lopez-ArmasAntioxid. Redox Signal., 13 (2010), pp. 193-247CrossRefView Record in Scopus[68]Z.T. Kelleher, A. Matsumoto, J.S. Stamler, H.E. MarshallJ. Biol. Chem., 282 (2007), pp. 30667-30672CrossRefView Record in Scopus[69]V. de Oliveira-Marques, L. Cyrne, H.S. Marinho, F. AntunesJ. Immunol., 178 (2007), pp. 3893-3902CrossRefView Record in Scopus[70]D.R. Clive, J.J. GreeneCell Biochem. Funct., 14 (1996), pp. 49-55CrossRefView Record in Scopus[71]E. Pineda-Molina, P. Klatt, J. Vazquez, A. Marina, M. Garcia de Lacoba, D. Perez-Sala, S. LamasBiochemistry, 40 (2001), pp. 14134-14142View Record in Scopus[72]J. Qin, G.M. Clore, W.M. Kennedy, J.R. Huth, A.M. GronenbornStructure, 3 (1995), pp. 289-297ArticleDownload PDFView Record in Scopus[73]N.L. Reynaert, A. van der Vliet, A.S. Guala, T. McGovern, M. Hristova, C. Pantano, N.H. Heintz, J. Heim, Y.S. Ho, D.E. Matthews, E.F. Wouters, Y.M. Janssen-HeiningerProc. Natl. Acad. Sci. U. S. A., 103 (2006), pp. 13086-13091CrossRefView Record in Scopus[74]R. Venugopal, A.K. JaiswalProc. Natl. Acad. Sci. U. S. A., 93 (1996), pp. 14960-14965View Record in Scopus[75]A.D. Kraft, D.A. Johnson, J.A. JohnsonJ. Neurosci., 24 (2004), pp. 1101-1112View Record in Scopus[76]J. Hirrlinger, R. DringenBrain Res. Rev., 63 (2010), pp. 177-188ArticleDownload PDFView Record in Scopus[77]A.A. Starkov, G. Fiskum, C. Chinopoulos, B.J. Lorenzo, S.E. Browne, M.S. Patel, M.F. BealJ. Neurosci., 24 (2004), pp. 7779-7788View Record in Scopus[78]A.L. Bulteau, M. Ikeda-Saito, L.I. SzwedaBiochemistry, 42 (2003), pp. 14846-14855View Record in Scopus[79]A.C. Nulton-Persson, L.I. SzwedaJ. Biol. Chem., 276 (2001), pp. 23357-23361View Record in Scopus[80]I. Oezen, W. Rossmanith, S. Forss-Petter, S. Kemp, T. Voigtlander, K. Moser-Thier, R.J. Wanders, R.E. Bittner, J. BergerHum. Mol. Genet., 14 (2005), pp. 1127-1137CrossRefView Record in Scopus[81]A. Herrero-Mendez, A. Almeida, E. Fernandez, C. Maestre, S. Moncada, J.P. BolanosNat. Cell Biol., 11 (2009), pp. 747-752CrossRefView Record in Scopus[82]J.E. Rinholm, N.B. Hamilton, N. Kessaris, W.D. Richardson, L.H. Bergersen, D. AttwellJ. Neurosci., 31 (2011), pp. 538-548CrossRefView Record in Scopus[83]J. Yao, R.W. Irwin, L. Zhao, J. Nilsen, R.T. Hamilton, R.D. BrintonProc. Natl. Acad. Sci. U. S. A., 106 (2009), pp. 14670-14675CrossRefView Record in Scopus[84]I. FerrerJ. Bioenerg. Biomembr., 41 (2009), pp. 425-431CrossRefView Record in Scopus[85]N. Cartier, S. Hacein-Bey-Abina, C.C. Bartholomae, G. Veres, M. Schmidt, I. Kutschera, M. Vidaud, U. Abel, L. Dal-Cortivo, L. Caccavelli, N. Mahlaoui, V. Kiermer, D. Mittelstaedt, C. Bellesme, N. Lahlou, F. Lefrere, S. Blanche, M. Audit, E. Payen, P. Leboulch, B. l’Homme, P. Bougneres, C. Von Kalle, A. Fischer, M. Cavazzana-Calvo, P. AubourgScience, 326 (2009), pp. 818-823CrossRefView Record in Scopus[86]G.S. HotamisligilNature, 444 (2006), pp. 860-867CrossRefView Record in Scopus[87]J.S. Henkel, J.I. Engelhardt, L. Siklos, E.P. Simpson, S.H. Kim, T. Pan, J.C. Goodman, T. Siddique, D.R. Beers, S.H. AppelAnn. Neurol., 55 (2004), pp. 221-235View Record in Scopus[88]J. Lopez-Erauskin, S. Fourcade, J. Galino, M. Ruiz, A. Schluter, A. Naudi, M. Jove, M. Portero-Otin, R. Pamplona, I. Ferrer, A. PujolAnn. Neurol., 70 (2011), pp. 84-92CrossRefView Record in Scopus[89]B. Halliwell, J.M.C. GutteridgeB. Halliwell, J.M.C. Gutteridge (Eds.), Free Radicals in Biology and Medicine, Clarendon Press, Oxford (1996), pp. 188-266Google Scholar[90]B.H. Harvey, C. Joubert, J.L. du Preez, M. BerkNeurochem. Res., 33 (2008), pp. 508-517CrossRefView Record in Scopus[91]P. Arivazhagan, C. PanneerselvamPharmacol. Res., 42 (2000), pp. 219-222ArticleDownload PDFView Record in Scopus[92]A. Ambrus, L. Tretter, V. Adam-ViziJ. Neurochem., 109 (Suppl. 1) (2009), pp. 222-229CrossRefView Record in Scopus[93]G. Lacraz, F. Figeac, J. Movassat, N. Kassis, J. Coulaud, A. Galinier, C. Leloup, D. Bailbe, F. Homo-Delarche, B. PorthaPLoS One, 4 (2009), p. e6500CrossRef[94]M.C. Rodriguez, J.R. MacDonald, D.J. Mahoney, G. Parise, M.F. Beal, M.A. TarnopolskyMuscle Nerve, 35 (2007), pp. 235-242CrossRefView Record in Scopus[95]J.S. Tan, J.J. Wang, V. Flood, E. Rochtchina, W. Smith, P. MitchellOphthalmology, 115 (2008), pp. 334-341ArticleDownload PDFView Record in Scopus[96]M.A. TarnopolskyAdv. Drug Deliv. Rev., 60 (2008), pp. 1561-1567ArticleDownload PDFView Record in Scopus[97]S. Fourcade, M. Ruiz, C. Camps, A. Schluter, S.M. Houten, P.A. Mooyer, T. Pampols, G. Dacremont, R.J. Wanders, M. Giros, A. PujolAm. J. Physiol. Endocrinol. Metab., 296 (2009), pp. E211-E221CrossRefView Record in Scopus[98]J. Liu, N.S. Sabeva, S. Bhatnagar, X.A. Li, A. Pujol, G.A. GrafJ. Lipid Res., 51 (2010), pp. 162-168CrossRefView Record in Scopus[99]J.F. Lu, E. Barron-Casella, R. Deering, A.K. Heinzer, A.B. Moser, K.L. deMesy, G.S. Bentley, C.M.M. Wand, Z. Pei, P.A. Watkins, A. Pujol, K.D. Smith, J.M. PowersLab. Invest., 87 (2007), pp. 261-272CrossRefView Record in Scopus[100]L.K. Kwong, S. Kamzalov, I. Rebrin, A.C. Bayne, C.K. Jana, P. Morris, M.J. Forster, R.S. SohalFree Radic. Biol. Med., 33 (2002), pp. 627-638ArticleDownload PDFView Record in Scopus[101]T.L. Wadsworth, J.A. Bishop, A.S. Pappu, R.L. Woltjer, J.F. QuinnJ. Alzheimers Dis., 14 (2008), pp. 225-234View Record in Scopus[102]M.P. MurphyExpert. Opin. Biol. Ther., 1 (2001), pp. 753-764CrossRefView Record in Scopus[103]R.A. Smith, G.F. Kelso, A.M. James, M.P. MurphyMethods Enzymol., 382 (2004), pp. 45-67ArticleDownload PDFView Record in Scopus[104]C. Lu, D. Zhang, M. Whiteman, J.S. ArmstrongAntioxid. Redox Signal., 10 (2008), pp. 651-660CrossRefView Record in Scopus[105]M.P. Murphy, R.A. SmithAnnu. Rev. Pharmacol. Toxicol., 47 (2007), pp. 629-656CrossRefView Record in Scopus[106]V. Calabrese, C. Cornelius, A.T. Dinkova-Kostova, E.J. Calabrese, M.P. MattsonAntioxid. Redox Signal., 13 (2010), pp. 1763-1811CrossRefView Record in Scopus[107]V. Calabrese, C. Cornelius, L. Maiolino, M. Luca, R. Chiaramonte, M.A. Toscano, A. SerraNeurochem. Res., 35 (2010), pp. 2208-2217CrossRefView Record in Scopus[108]F. Di Domenico, M. Perluigi, D.A. Butterfield, C. Cornelius, V. CalabreseNeurochem. Res., 35 (2010), pp. 2184-2192CrossRefView Record in Scopus[109]S. Hauptmann, I. Scherping, S. Drose, U. Brandt, K.L. Schulz, M. Jendrach, K. Leuner, A. Eckert, W.E. MullerNeurobiol. Aging, 30 (2009), pp. 1574-1586ArticleDownload PDFView Record in Scopus[110]K. Leuner, S. Hauptmann, R. Abdel-Kader, I. Scherping, U. Keil, J.B. Strosznajder, A. Eckert, W.E. MullerAntioxid. Redox Signal., 9 (2007), pp. 1659-1675View Record in Scopus[111]A. Martinez, M. Portero-Otin, R. Pamplona, I. FerrerBrain Pathol., 20 (2010), pp. 281-297CrossRefView Record in Scopus[112]K.M. Humphries, L.I. SzwedaBiochemistry, 37 (1998), pp. 15835-15841View Record in Scopus[113]L. Tretter, V. Adam-ViziPhilos. Trans. R. Soc. Lond. B Biol. Sci., 360 (2005), pp. 2335-2345CrossRefView Record in Scopus[114]W. YingAntioxid. Redox Signal., 10 (2008), pp. 179-206CrossRefView Record in Scopus[115]G.E. Gibson, A. Starkov, J.P. Blass, R.R. Ratan, M.F. BealBiochim. Biophys. Acta, 1802 (2010), pp. 122-134ArticleDownload PDFView Record in Scopus[116]V. Rhein, X. Song, A. Wiesner, L.M. Ittner, G. Baysang, F. Meier, L. Ozmen, H. Bluethmann, S. Drose, U. Brandt, E. Savaskan, C. Czech, J. Gotz, A. EckertProc. Natl. Acad. Sci. U. S. A., 106 (2009), pp. 20057-20062CrossRefView Record in Scopus[117]S. Gispert, F. Ricciardi, A. Kurz, M. Azizov, H.H. Hoepken, D. Becker, W. Voos, K. Leuner, W.E. Muller, A.P. Kudin, W.S. Kunz, A. Zimmermann, J. Roeper, D. Wenzel, M. Jendrach, M. Garcia-Arencibia, J. Fernandez-Ruiz, L. Huber, H. Rohrer, M. Barrera, A.S. Reichert, U. Rub, A. Chen, R.L. Nussbaum, G. AuburgerPLoS One, 4 (2009), p. e5777CrossRef[118]K.J. De Vos, A.J. Grierson, S. Ackerley, C.C. MillerAnnu. Rev. Neurosci., 31 (2008), pp. 151-173CrossRefView Record in Scopus[119]G.B. Stokin, C. Lillo, T.L. Falzone, R.G. Brusch, E. Rockenstein, S.L. Mount, R. Raman, P. Davies, E. Masliah, D.S. Williams, L.S. GoldsteinScience, 307 (2005), pp. 1282-1288CrossRefView Record in Scopus[120]O. Lazarov, G.A. Morfini, G. Pigino, A. Gadadhar, X. Chen, J. Robinson, H. Ho, S.T. Brady, S.S. SisodiaJ. Neurosci., 27 (2007), pp. 7011-7020CrossRefView Record in Scopus[121]C.R. Jack Jr., H.J. Wiste, P. Vemuri, S.D. Weigand, M.L. Senjem, G. Zeng, M.A. Bernstein, J.L. Gunter, V.S. Pankratz, P.S. Aisen, M.W. Weiner, R.C. Petersen, L.M. Shaw, J.Q. Trojanowski, D.S. KnopmanBrain, 133 (2010), pp. 3336-3348CrossRefView Record in Scopus[122]R.A. Sperling, P.S. Aisen, L.A. Beckett, D.A. Bennett, S. Craft, A.M. Fagan, T. Iwatsubo, C.R. Jack Jr., J. Kaye, T.J. Montine, D.C. Park, E.M. Reiman, C.C. Rowe, E. Siemers, Y. Stern, K. Yaffe, M.C. Carrillo, B. Thies, M. Morrison-Bogorad, M.V. Wagster, C.H. PhelpsAlzheimers Dement., 7 (2011), pp. 280-292ArticleDownload PDFCrossRefView Record in Scopus[123]W.R. Markesbery, M.A. LovellAntioxid. Redox Signal., 8 (2006), pp. 2039-2045CrossRefView Record in Scopus[124]R. Pamplona, E. Dalfo, V. Ayala, M.J. Bellmunt, J. Prat, I. Ferrer, M. Portero-OtinJ. Biol. Chem., 280 (2005), pp. 21522-21530View Record in Scopus[125]T. Reed, M. Perluigi, R. Sultana, W.M. Pierce, J.B. Klein, D.M. Turner, R. Coccia, W.R. Markesbery, D.A. ButterfieldNeurobiol. Dis., 30 (2008), pp. 107-120ArticleDownload PDFView Record in Scopus[126]R. Sultana, D. Boyd-Kimball, H.F. Poon, J. Cai, W.M. Pierce, J.B. Klein, M. Merchant, W.R. Markesbery, D.A. ButterfieldNeurobiol. Aging, 27 (2006), pp. 1564-1576ArticleDownload PDFView Record in Scopus[127]K. Hirai, G. Aliev, A. Nunomura, H. Fujioka, R.L. Russell, C.S. Atwood, A.B. Johnson, Y. Kress, H.V. Vinters, M. Tabaton, S. Shimohama, A.D. Cash, S.L. Siedlak, P.L. Harris, P.K. Jones, R.B. Petersen, G. Perry, M.A. SmithJ. Neurosci., 21 (2001), pp. 3017-3023View Record in Scopus[128]X. Wang, B. Su, H.G. Lee, X. Li, G. Perry, M.A. Smith, X. ZhuJ. Neurosci., 29 (2009), pp. 9090-9103CrossRefView Record in Scopus[129]M. Dumont, M.F. BealFree Radic. Biol. Med., 51 (2011), pp. 1014-1026ArticleDownload PDFView Record in Scopus[130]X. Yang, Y. Yang, G. Li, J. Wang, E.S. YangJ. Mol. Neurosci., 34 (2008), pp. 165-171CrossRefView Record in Scopus[131]X. Yang, G. Dai, G. Li, E.S. YangJ. Mol. Neurosci., 41 (2010), pp. 110-113CrossRefView Record in Scopus[132]S. Tucker, M. Ahl, A. Bush, D. Westaway, X. Huang, J.T. RogersCurr. Alzheimer Res., 2 (2005), pp. 249-254CrossRefView Record in Scopus[133]A.L. Fu, Z.H. Dong, M.J. SunBrain Res., 1109 (2006), pp. 201-206ArticleDownload PDFView Record in Scopus[134]A. Chan, J. Paskavitz, R. Remington, S. Rasmussen, T.B. SheaAm. J. Alzheimers Dis. Other Demen., 23 (2008), pp. 571-585View Record in Scopus[135]R. Remington, A. Chan, J. Paskavitz, T.B. SheaAm. J. Alzheimers Dis. Other Demen., 24 (2009), pp. 27-33CrossRefView Record in Scopus[136]V.H. Perry, J.A. Nicoll, C. HolmesNat. Rev. Neurol., 6 (2010), pp. 193-201CrossRefView Record in Scopus[137]M.P. Lichtenstein, P. Carriba, R. Masgrau, A. Pujol, E. GaleaFront Aging Neurosci., 2 (2010), p. 142[138]C.S. Casley, L. Canevari, J.M. Land, J.B. Clark, M.A. SharpeJ. Neurochem., 80 (2002), pp. 91-100View Record in Scopus[139]C.S. Casley, J.M. Land, M.A. Sharpe, J.B. Clark, M.R. Duchen, L. CanevariNeurobiol. Dis., 10 (2002), pp. 258-267ArticleDownload PDFView Record in Scopus[140]I.V. Murray, M.E. Sindoni, P.H. AxelsenBiochemistry, 44 (2005), pp. 12606-12613CrossRefView Record in Scopus[141]J. Vina, A. Lloret, E. Giraldo, M.C. Badia, M.D. AlonsoCurr. Pharm. Des., 17 (2011), pp. 3861-3864CrossRefView Record in Scopus[142]H. Misonou, M. Morishima-Kawashima, Y. IharaBiochemistry, 39 (2000), pp. 6951-6959View Record in Scopus[143]H. Akhter, A. Katre, L. Li, X. Liu, R.M. LiuJ. Alzheimers Dis. (2011)[144]E. Tamagno, M. Parola, P. Bardini, A. Piccini, R. Borghi, M. Guglielmotto, G. Santoro, A. Davit, O. Danni, M.A. Smith, G. Perry, M. TabatonJ. Neurochem., 92 (2005), pp. 628-636CrossRefView Record in Scopus[145]F. Li, N.Y. Calingasan, F. Yu, W.M. Mauck, M. Toidze, C.G. Almeida, R.H. Takahashi, G.A. Carlson, M. Flint Beal, M.T. Lin, G.K. GourasJ. Neurochem., 89 (2004), pp. 1308-1312View Record in Scopus[146]C.A. Hansson, S. Frykman, M.R. Farmery, L.O. Tjernberg, C. Nilsberth, S.E. Pursglove, A. Ito, B. Winblad, R.F. Cowburn, J. Thyberg, M. AnkarcronaJ. Biol. Chem., 279 (2004), pp. 51654-51660View Record in Scopus[147]M. Soler-Lopez, A. Zanzoni, R. Lluis, U. Stelzl, P. AloyGenome Res., 21 (2011), pp. 364-376CrossRefView Record in Scopus[148]C.R. Jack Jr., V.J. Lowe, S.D. Weigand, H.J. Wiste, M.L. Senjem, D.S. Knopman, M.M. Shiung, J.L. Gunter, B.F. Boeve, B.J. Kemp, M. Weiner, R.C. PetersenBrain, 132 (2009), pp. 1355-1365CrossRefView Record in Scopus[149]B. Terni, J. Boada, M. Portero-Otin, R. Pamplona, I. FerrerBrain Pathol., 20 (2010), pp. 222-233CrossRefView Record in Scopus[150]L. Mosconi, V. Berti, L. Glodzik, A. Pupi, S. De Santi, M.J. de LeonJ. Alzheimers Dis., 20 (2010), pp. 843-854View Record in Scopus[151]D. Pratico, K. Uryu, S. Leight, J.Q. Trojanoswki, V.M. LeeJ. Neurosci., 21 (2001), pp. 4183-4187View Record in Scopus[152]P.H. Reddy, S. McWeeney, B.S. Park, M. Manczak, R.V. Gutala, D. Partovi, Y. Jung, V. Yau, R. Searles, M. Mori, J. QuinnHum. Mol. Genet., 13 (2004), pp. 1225-1240View Record in Scopus[153]D. Galasko, T.J. MontineBiomark. Med., 4 (2010), pp. 27-36CrossRefView Record in Scopus[154]C.B. Do, J.Y. Tung, E. Dorfman, A.K. Kiefer, E.M. Drabant, U. Francke, J.L. Mountain, S.M. Goldman, C.M. Tanner, J.W. Langston, A. Wojcicki, N. ErikssonPLoS Genet., 7 (2011), p. e1002141CrossRef[155]G. Morfini, G. Pigino, K. Opalach, Y. Serulle, J.E. Moreira, M. Sugimori, R.R. Llinas, S.T. BradyProc. Natl. Acad. Sci. U. S. A., 104 (2007), pp. 2442-2447CrossRefView Record in Scopus[156]T.K. Lin, C.W. Liou, S.D. Chen, Y.C. Chuang, M.M. Tiao, P.W. Wang, J.B. Chen, J.H. ChuangChang Gung Med. J., 32 (2009), pp. 589-599View Record in Scopus[157]R. Banerjee, A.A. Starkov, M.F. Beal, B. ThomasBiochim. Biophys. Acta, 1792 (2009), pp. 651-663ArticleDownload PDFView Record in Scopus[158]D.J. Surmeier, J.N. Guzman, J. Sanchez-PadillaCell Calcium, 47 (2010), pp. 175-182ArticleDownload PDFView Record in Scopus[159]C.A. Gautier, T. Kitada, J. ShenProc. Natl. Acad. Sci. U. S. A., 105 (2008), pp. 11364-11369CrossRefView Record in Scopus[160]H. Shimura, N. Hattori, S. Kubo, Y. Mizuno, S. Asakawa, S. Minoshima, N. Shimizu, K. Iwai, T. Chiba, K. Tanaka, T. SuzukiNat. Genet., 25 (2000), pp. 302-305View Record in Scopus[161]L.S. Chin, J.A. Olzmann, L. LiBiochem. Soc. Trans., 38 (2010), pp. 144-149CrossRefView Record in Scopus[162]D.C. Angeles, B.H. Gan, L. Onstead, Y. Zhao, K.L. Lim, J. Dachsel, H. Melrose, M. Farrer, Z.K. Wszolek, D.W. Dickson, E.K. TanHum. Mutat. (2011)[163]K. Banerjee, M. Sinha, L. Pham Cle, S. Jana, D. Chanda, R. Cappai, S. ChakrabartiFEBS Lett., 584 (2010), pp. 1571-1576ArticleDownload PDFCrossRefView Record in Scopus[164]F. Kamp, N. Exner, A.K. Lutz, N. Wender, J. Hegermann, B. Brunner, B. Nuscher, T. Bartels, A. Giese, K. Beyer, S. Eimer, K.F. Winklhofer, C. HaassEmbo J., 29 (2010), pp. 3571-3589CrossRefView Record in Scopus[165]B. Thomas, R. Banerjee, N.N. Starkova, S.F. Zhang, N.Y. Calingasan, L. Yang, E. Wille, B.J. Lorenzo, D.J. Ho, M.F. Beal and A. Starkov, Antioxid. Redox Signal. (2011).Google Scholar[166]K.E. Miller, M.P. SheetzJ. Cell Sci., 117 (2004), pp. 2791-2804View Record in Scopus[167]A.H. Schapira, J.M. Cooper, D. Dexter, P. Jenner, J.B. Clark, C.D. MarsdenLancet, 1 (1989), p. 1269ArticleDownload PDFView Record in Scopus[168]H.F. Poon, M. Frasier, N. Shreve, V. Calabrese, B. Wolozin, D.A. ButterfieldNeurobiol. Dis., 18 (2005), pp. 492-498View Record in Scopus[169]P.M. Keeney, J. Xie, R.A. Capaldi, J.P. Bennett Jr.J. Neurosci., 26 (2006), pp. 5256-5264CrossRefView Record in Scopus[170]E. Dalfo, I. FerrerNeurobiol. Aging, 29 (2008), pp. 408-417ArticleDownload PDFView Record in Scopus[171]D. Yao, Z. Gu, T. Nakamura, Z.Q. Shi, Y. Ma, B. Gaston, L.A. Palmer, E.M. Rockenstein, Z. Zhang, E. Masliah, T. Uehara, S.A. LiptonProc. Natl. Acad. Sci. U. S. A., 101 (2004), pp. 10810-10814View Record in Scopus[172]J. Choi, M.C. Sullards, J.A. Olzmann, H.D. Rees, S.T. Weintraub, D.E. Bostwick, M. Gearing, A.I. Levey, L.S. Chin, L. LiJ. Biol. Chem., 281 (2006), pp. 10816-10824CrossRefView Record in Scopus[173]E. Dalfo, M. Portero-Otin, V. Ayala, A. Martinez, R. Pamplona, I. FerrerJ. Neuropathol. Exp. Neurol., 64 (2005), pp. 816-830CrossRefView Record in Scopus[174]R. Savica, N. Kumar, J.E. Ahlskog, K.A. Josephs, J.Y. Matsumoto, A. McKeonParkinsonism Relat. Disord, 18 (2012), pp. 30-34ArticleDownload PDFView Record in Scopus[175]I. FerrerPark. Dis. (2011), p. 708404View Record in Scopus[176]E. Minones-Moyano, S. Porta, G. Escaramis, R. Rabionet, S. Iraola, B. Kagerbauer, Y. Espinosa-Parrilla, I. Ferrer, X. Estivill, E. MartiHum. Mol. Genet., 20 (2011), pp. 3067-3078CrossRefView Record in Scopus[177]Z.D. Zhou, S.Y. Kerk, G.G. Xiong, T.M. LimJ. Neurochem., 108 (2009), pp. 601-610CrossRefView Record in Scopus[178]K.A. Bruggink, H.B. Kuiperij, F. Ekholm-Pettersson, M.M. VerbeekNeurology, 77 (2011), p. 510author reply 510–1CrossRefView Record in Scopus[179]U. Shirendeb, A.P. Reddy, M. Manczak, M.J. Calkins, P. Mao, D.A. Tagle, P.H. ReddyHum. Mol. Genet., 20 (2011), pp. 1438-1455CrossRefView Record in Scopus[180]C. Turner, A.H. SchapiraJ. Bioenerg. Biomembr., 42 (2010), pp. 193-198CrossRefView Record in Scopus[181]A. Wyttenbach, O. Sauvageot, J. Carmichael, C. Diaz-Latoud, A.P. Arrigo, D.C. RubinszteinHum. Mol. Genet., 11 (2002), pp. 1137-1151View Record in Scopus[182]M. Gu, M.T. Gash, V.M. Mann, F. Javoy-Agid, J.M. Cooper, A.H. SchapiraAnn. Neurol., 39 (1996), pp. 385-389CrossRefView Record in Scopus[183]W. Song, J. Chen, A. Petrilli, G. Liot, E. Klinglmayr, Y. Zhou, P. Poquiz, J. Tjong, M.A. Pouladi, M.R. Hayden, E. Masliah, M. Ellisman, I. Rouiller, R. Schwarzenbacher, B. Bossy, G. Perkins, E. Bossy-Wetzel NatMed., 17 (2011), pp. 377-382CrossRef[184]N. Nakao, P. BrundinNeuroscience, 76 (1997), pp. 749-761[185]M.A. Sorolla, M.J. Rodriguez-Colman, J. Tamarit, Z. Ortega, J.J. Lucas, I. Ferrer, J. Ros, E. CabiscolFree Radic Biol Med, 49 (2010), pp. 612-621ArticleDownload PDFView Record in Scopus[186]I. Tunez, F. Sanchez-Lopez, E. Aguera, R. Fernandez-Bolanos, F.M. Sanchez, I. Tasset-CuevasJ. Med. Chem., 54 (2011), pp. 5602-5606CrossRefView Record in Scopus[187]G. Schilling, M.L. Coonfield, C.A. Ross, D.R. BorcheltNeurosci. Lett., 315 (2001), pp. 149-153ArticleDownload PDFView Record in Scopus[188]E. Vamos, K. Voros, L. Vecsei, P. KlivenyiBiomed. Pharmacother., 64 (2010), pp. 282-286ArticleDownload PDFView Record in Scopus[189]C. Stack, D. Ho, E. Wille, N.Y. Calingasan, C. Williams, K. Liby, M. Sporn, M. Dumont, M.F. BealFree Radic. Biol. Med., 49 (2010), pp. 147-158ArticleDownload PDFView Record in Scopus[190]D.W. Weir, A. Sturrock and B.R. Leavitt, Lancet Neurol. 10 573–90.Google Scholar[191]J. Lin, P.H. Wu, P.T. Tarr, K.S. Lindenberg, J. St-Pierre, C.Y. Zhang, V.K. Mootha, S. Jager, C.R. Vianna, R.M. Reznick, L. Cui, M. Manieri, M.X. Donovan, Z. Wu, M.P. Cooper, M.C. Fan, L.M. Rohas, A.M. Zavacki, S. Cinti, G.I. Shulman, B.B. Lowell, D. Krainc, B.M. SpiegelmanCell, 119 (2004), pp. 121-135ArticleDownload PDFView Record in Scopus