REVIEWS

NEURODEGENERATIVE DISEASES
AND OXIDATIVE STRESS
Kevin J. Barnham*, Colin L. Masters* and Ashley I. Bush*‡
Oxidative stress has been implicated in the progression of Alzheimer’s disease, Parkinson’s
disease and amyotrophic lateral sclerosis. Oxygen is vital for life but is also potentially dangerous,
and a complex system of checks and balances exists for utilizing this essential element. Oxidative
stress is the result of an imbalance in pro-oxidant/antioxidant homeostasis that leads to the
generation of toxic reactive oxygen species. The systems in place to cope with the biochemistry
of oxygen are complex, and many questions about the mechanisms of oxygen regulation remain
unanswered. However, this same complexity provides a number of therapeutic targets, and
different strategies, including novel metal–protein attenuating compounds, aimed at a variety of
targets have shown promise in clinical studies.

HIPPOCAMPUS
A region of the brain consisting
of the grey matter at the bottom
of the lateral ventricle that is
involved in motivation, emotion
and the formation of memory.
*Department of Pathology,
The University of Melbourne,
and The Mental Health
Research Institute of Victoria,
Victoria 3010, Australia.
‡
Laboratory for Oxidation
Biology, Genetics and Aging
Research Unit and
Department of Psychiatry,
Harvard Medical School,
Massachusetts General
Hospital East, Charlestown,
Massachusetts 02129, USA.
Correspondence to A.I.B.
e-mail: bush@helix.mgh.
harvard.edu
doi:10.1038/nrd1330
Neurodegenerative diseases, such as Alzheimer’s disease
(AD), Parkinson’s disease (PD) and amyotrophic lateral
sclerosis (ALS), are defined by the progressive loss of
specific neuronal cell populations and are associated
with protein aggregates. A common feature of these
diseases is extensive evidence of oxidative stress, which
might be responsible for the dysfunction or death of
neuronal cells that contributes to disease pathogenesis.
Oxidative stress is the result of unregulated production
of reactive oxygen species (ROS), such as hydrogen per-
oxide, nitric oxide, superoxide and the highly reactive
hydroxyl radicals. High oxygen consumption, relatively
low antioxidant levels and low regenerative capacity
result in brain tissue being susceptible to oxidative
damage. Although a contentious idea, many believe that
oxidative stress contributes to the neurodegeneration
associated with these diseases; as a result, a variety of
therapeutic strategies target this phenomena.
Evidence of oxidative stress
Unsaturated lipids are particularly susceptible to oxida-
tive modification and lipid peroxidation is a sensitive
marker of oxidative stress. Lipid peroxidation is the
result of attack by radicals on the double bond of unsatu-
rated fatty acids, such as linoleic acid and arachidonic
acid, to generate highly reactive lipid peroxy radicals
that initiate a chain reaction of further attacks on other
unsaturated fatty acids. The chain reaction leads to the
formation of breakdown products including 4-hydroxy-
2,3-nonenal (HNE), acrolein, malondialdehyde and
F2-isoprostanes. Elevated HNE levels have been observed
in AD1,2 and PD3 brain tissue, whereas increased HNE
has been observed in the cerebrospinal fluid (CSF) of
ALS patients4. Acrolein, thiobarbituric acid-reactive
substances (TBARs, the most prevalent substrate of
which is malondialdehyde) and F2-isoprostanes are
all increased in AD brains relative to age-matched
controls5. Malondialdehyde is increased in PD brains3,
whereas increased TBARs have been observed in the
plasma of ALS patients6.
All four bases of DNA are susceptible to oxidative
damage involving hydroxylation7, and significant pro-
tein carbonylation (for example, of creatine kinase and
β-actin2) and nitration is observed in AD brains,
whereas increased levels of 8-hydroxyguanine and
8-hydroxy-2-deoxyguanosine are observed in PD brains;
the selective attack on guanine bases implies OH radicals
as the oxidative species8.
Cells have endogenous defence mechanisms against
oxidative stress and changes in these are also used as
markers of oxidative stress. In the AD brain, the activity
of the antioxidant proteins catalase, superoxide dismutase
(SOD), glutathione peroxidase and glutathione reductase
are increased in the HIPPOCAMPUS and amygdala9,10.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 3 | MARCH 2004 | 2 0 5

REVIEWS
Box 1 | Friedreich’s ataxia
An abnormal GAA trinucleotide expansion within the first intron of the gene encoding
frataxin, a mitochondrial protein, leads to a deficiency of frataxin and is the major cause of
Friedreich’s ataxia (FA)118. One consequence of this deficiency is iron accumulation within
the mitochondria, which leads to increased oxidative stress resulting in cardiomyopathy
and neurodegeneration. Another consequence of the iron overload is a breakdown in the
mitochondrial respiratory chain due to a deficiency in Fe-S enzymes119. Two therapeutic
strategies are now being investigated that target the Fe-mediated oxidative stress. The first
involves idebenone, a free-radical scavenger analogue of coenzyme Q10, although results
from trials with this compound have been mixed. Idebenone has been shown to reduce
oxidative damage (as measured by urinary 8-hydroxy-2’-deoxyguanosine levels) and
rescue respiratory chain function120. Several pilot clinical studies have shown that
idebenone has significant benefit with respect to the hypertrophic cardiomyopathy
associated with FA; however, there were no indications of improved neurological
functions120–122. An alternative strategy is to target the excess Fe3+ via the use of metal
chelators. Fibroblasts isolated from FA patients are particularly susceptible to oxidative
insults, and in a cell-culture model using these cells the iron chelator desferrioxamine
(DFO) was found to be protective123. The iron overload associated with FA is a specific
overload of the mitochondria, not a gross overload; chelation strategies therefore need
to target this iron specifically. DFO has poor bioavailability and is unlikely to have a
significant effect on the iron levels within the mitochondria. A new class of iron
chelators based on 2-pyridylcarboxaldehyde isonicotinoyl hydrazones that can target
mitochondrial iron has recently been developed, but the applicability of these molecules
as potential therapeutics for FA is still being assessed124.
How do ROS kill cells?
As their name suggests, ROS are very reactive and will
react with a multitude of different molecules to initiate
neuronal cell death and hence neurodegeneration
through an array of different pathways. Oxidative stress
(that is, oxidation of lipids, proteins and DNA) results in
impaired cellular functions and the formation of toxic
species, such as peroxides, alcohols, aldehydes, ketones
and cholesterol oxide. Cholesterol oxide is toxic to
lymphocytes and blood vessel macrophages11.
The oxidatively modified lipids acrolein and HNE
induce toxicity by crosslinking to cystine, lysine and
histidine residues via a Michael addition. Acrolein down-
regulates the uptake of glutamate and glucose from cell
culture12, whereas HNE modifies proteins resulting in a
multitude of effects, including inhibition of the neuronal
glucose transporter type-3, the glutamate transporter
GLT-1 (REF. 13), as well as the Na++K+ ATPases14. HNE
activates c-Jun aminoterminal kinases and mitogen-
activated protein kinase-1 (also known as p38), thereby
stimulating an apoptotic cascade15. Modifications to
proteins result in the impairment of enzymes (for
example, glutamine synthase, superoxide dismutase),
whereas ROS interactions with DNA lead to mutations.
The excessive generation of ROS leads to dysregula-
tion of intracellular calcium signalling, and such dysreg-
ulation has been widely observed in neurodegenerative
REDOX
diseases in which aberrant calcium levels stimulate
A reversible chemical reaction multiple pathways that ultimately induce an apoptotic
in which one reaction is an cascade (reviewed in REFS 16–19).
oxidation reaction and the One of the downstream events that occurs in
reverse a reduction.
response to an ROS-induced calcium influx is an excito-
FENTON REACTION toxic response19,20, that is, activation of glutamate recep-
Mn+ + H2O2 → M(n+1)+ + OH– tors that trigger a cascade of events leading to cell death.
+ OH•. Excitotoxic responses have been implicated in several
206 | MARCH 2004 | VOLUME 3
neurological conditions, including epilepsy and stroke,
as well as the neurodegenerative diseases AD, PD, ALS
and Huntington’s disease (reviewed in REF. 21).
How are ROS generated?
The generation of ROS requires the activation of molecu-
lar oxygen. At the quantum level, O2 is in a triplet spin-
state and consequently interactions with most organic
molecules are spin-forbidden (that is, they are energeti-
cally unfavourable). As utilization of O2 is a prerequisite
for most life forms, this intrinsic feature of O2 must be
overcome. Organisms have evolved a range of metallo-
enzymes to take advantage of the interactions between
O2 and metal ions to activate molecular oxygen as ROS;
the subsequent free radicals are an intrinsic part of
normal metabolism. As ROS are also toxic, cells have
developed highly elaborate means of regulating both
metal ion interactions and the generation of ROS. The
same properties that cells harness for beneficial means
become destructive when the regulatory processes
breakdown.
ROS are produced by a number of different pathways,
including direct interactions between REDOX-active metals
and oxygen species via reactions such as the FENTON and
Haber–Weiss reactions, or via indirect pathways involving
the calcium activation of metallo-enzymes such as
phospholipases, nitric oxide synthase and xanthine
dehydrogenase (also known as xanthine oxidase)22.
Calcium is crucial to signal transduction and as such is
both sensitive to a number of different stimuli and able
to elicit a variety of different cellular responses. There is a
large body of evidence documenting disruption of
calcium homeostasis in neurodegenerative diseases,
leading to a breakdown in a large number of cellular
processes. However, most of these different signalling
pathways rely on feedback signalling and it is conse-
quently difficult to differentiate cause and effect. An
example of this is the interplay between calcium sig-
nalling and ROS generation: although increases in
intracellular calcium have been reported to induce the
production of ROS22, at subtoxic levels ROS have been
shown to be important cell signallers and will induce an
increase in cytosolic calcium23,24. This has lead to much
debate as to whether the observed disruption of calcium
homeostasis is a cause or consequence of ROS generation.
As a general principle, the chemical origin of the
majority of ROS is the reaction of molecular oxygen
with the redox-active metals copper and iron25. The
ability of these metal ions to occupy multiple valence
states and undertake facile redox cycling, thereby acti-
vating molecular oxygen, has been utilized by a variety
of enzymes. However, unregulated redox-active metals
will inappropriately react with oxygen to generate ROS.
We have proposed that the proteins implicated in several
age-dependent neurodegenerative disorders (Aβ in AD,
α-synuclein in PD, SOD1 in ALS, frataxin in Friedreich’s
ataxia (BOX 1) and α-B-crystallin in cataracts), might
abnormally present Cu2+ or Fe3+ ligands for inappropriate
reaction with O2 (FIG. 1). These proteins might have some
aspect of their function subserved by these metal ions,
which normally occupy higher-affinity, embedded,
www.nature.com/reviews/drugdisc

O2 O2–/H2O2
Catechols
Vitamin C e–
Cholesterol, Fe2+/Cu+
arachidonic Cu+ Cu2+
acid
and so on
Drugs OH•
Aβ Lipid peroxidation
SOD Protein oxidation
α-synuclein Protein aggregation
α-crystallin DNA/RNA adducts
Figure 1 | ROS generation by abnormal reaction of O2 with protein-bound Fe or Cu. One of
the consequences of normal aging is that the levels of the redox-active metals copper and iron
in the brain increase. This increase could lead to hypermetallation of proteins that normally bind
redox-active metals at shielded sites. Adventitial binding — for example; at a loading site — will
increase the likelihood that ROS are generated inappropriately as illustrated, leading to the
oxidative stress that is observed in neurodegenerative diseases. Aβ, amyloid-β; ROS, reactive
oxygen species; SOD, superoxide dismutase.
redox-shielded binding sites. As metal concentration
rises in the brain with age, the probability increases that
a redox-competent, low-affinity metal-binding site will
recruit a metal ion from the normally redox-silent cellular
pool. In this manner, proteins such as Aβ can harness
endogenous biometals to foster the release of inappro-
priate redox activity and ROS generation.
Metal ions and neurodegenerative diseases
The dominant risk factor associated with the neuro-
degenerative diseases is increasing age. Several studies in
mice have shown that one of the consequences of normal
aging is a rise in the levels of copper and iron in brain
tissue (FIG. 1)26–28. The brain is an organ that concentrates
metal ions and recent evidence suggests that a break-
down in metal homeostasis is a key factor in a variety of
age-related neurodegenerative diseases29,30.
Alzheimer’s disease. Genetic evidence from cases of
familial AD indicates that Aβ metabolism is linked to
the disease31,32. AD is characterized by the deposition
of AMYLOID plaques, the major constituent being the amy-
loid-β peptide (Aβ) that is cleaved from the membrane-
bound amyloid precursor protein (APP)33–35. Although
the function of APP is unknown, recent evidence suggests
it functions in maintaining copper homeostasis (BOX 2).
Interactions of Aβ and metals. Recent results have high-
lighted the importance of Zn2+ in amyloid plaque for-
mation; for example, age- and female-sex-related plaque
AMYLOID
Protein/peptide deposited formation in Tg2576 transgenic mice was reduced by
in diseased tissue, with high genetic ablation of the zinc transporter 3 protein, which
β-sheet structure. is required for zinc transport into synaptic vesicles36.
The plaques could be described as metallic sinks
NEUROPILE
The mass of closely packed nerve
because remarkably high concentrations of Cu (400
cell processes comprising the µM), Zn (1 mM) and Fe (1 mM) have been found
central part of a ganglion. within the amyloid deposits in AD-affected brains37,38.
NATURE REVIEWS | DRUG DISCOVERY
REVIEWS
In vitro studies have shown that low µM levels of Zn2+
will induce protease-resistant aggregation and precipita-
tion of Aβ39. Cu2+ and Fe3+ also induce peptide aggrega-
tion that is exaggerated at acidic pH40. These metals are
normally found at high concentrations in the region of
the brain most susceptible to AD neurodegeneration.
During neurotransmission, high concentrations of Zn
(300 µM) and Cu (30 µM) are released, which might
explain why Aβ precipitation into amyloid commences
in the synapse36,41.
The cause of the neuronal cell loss in AD might be
related to oxidative stress from excessive free-radical
generation29,30,38,42,43. We have proposed that the major
source of oxidative stress and free-radical production in
the brain in AD is the transition metals Cu and Fe29,43
when bound to Aβ.
There is a large body of evidence indicating that the
homeostasis of Zn, Cu and Fe, and their respective
binding proteins, are significantly altered in the AD
brain (reviewed in REF. 44). A microparticle-induced
X-ray emission analysis of the cortical and accessory
basal nuclei of the amygdala indicated that these metals
accumulate in the NEUROPILE of the AD brain in concen-
trations that are three- to fivefold increased compared
with age-matched controls37. Evidence for abnormal Cu
homeostasis in AD includes a 2.2-times increase in the
concentration of Cu in CSF45, and an accompanying
increase of plasma Cu in AD46. There is an extensive lit-
erature describing abnormal levels of Fe and Fe-binding
proteins in AD47. Notably, the Fe that is found within
the amyloid deposits of human brain and in amyloid-
bearing APP transgenic mice is redox-active38,48. Raman
spectroscopy studies have demonstrated that Zn2+ and
Cu2+ are coordinated to the histidine residues of the
deposited Aβ in the senile plaque cores from diseased
brain tissue and that the sulphur atom of methionine 35
of Aβ is oxidized, which is indicative of a pro-oxidant
environment49.
Toxic mechanism(s) of Aβ. Synthetic Aβ is toxic to cells
in the presence of Cu2+, but this toxicity is inhibited by
extracellular catalase, which implicates H2O2 in the toxic
pathway50,51. When Cu2+ or Fe3+ coordinate Aβ, exten-
sive redox chemical reactions take place that reduce the
oxidation state of both metals and produce H2O2 from
O2 in a catalytic manner51–54. The formal reduction
potential of Cu2+ to Cu+ by Aβ42 is highly positive
(approximately +500–550 mV versus Ag/AgCl) and
characteristic of strongly reducing cupro-proteins52.
The generation of H2O2 in the presence of the reduced
form of the metal creates conditions in which Fenton
chemistry occurs with the generation of highly toxic
OH• radicals53.
The toxicity of Aβ to neuronal cells is enhanced by
the presence of Cu, which is usually present in culture
media51. Aβ possesses histidine residues at positions 6,
13 and 14, which form a structural element that enables
Aβ to coordinate transition metal ions. A variety of
spectroscopic studies have confirmed that the histidine
residues constitute the principal site(s) of metal coordi-
nation. The interactions of synthetic Aβ with copper,
VOLUME 3 | MARCH 2004 | 2 0 7
REVIEWS

Box 2 | Amyloid precursor protein and copper homeostasis

To prevent transition-metal-mediated oxidative stress, cells have
evolved complex metal-transport systems that deliver Cu and
Fe to metallo-enzymes and proteins. These include the Cu
chaperone for superoxide dismutase 1 (SOD1), which facilitates
insertion of Cu into the active site of SOD125, and the Cu ATPase
that is mutated in Wilson’s disease126. The chaperones direct the
Cu atoms to specific intracellular proteins, which essentially
results in an absence of unbound Cu in the intracellular
environment127. Therefore, cuproproteins have an important
role in maintaining cellular Cu metabolism128. Amyloid
precursor protein (APP) is a ubiquitously expressed, high-
turnover protein and recent data suggest that APP and Aβ could
have some role in metal homeostasis. Utilizing transgenic mouse models it has been shown that overexpression of the
carboxyl-terminal fragment of APP, which contains Aβ, results in significantly reduced copper and iron levels in transgenic
mouse brain; overexpression of APP in Tg2576 transgenic mice results in significantly reduced copper, but not iron28,
whereas APP knockout mice have increased copper levels in the brain and liver129. Copper levels can modulate APP
processing130,131, with higher copper levels resulting in a reduction in Aβ production and a consequential increase in the
non-amyloidogenic p3 form of the peptide132. The effect of increasing brain Cu levels in reducing cerebral Aβ burden in
APP transgenic mice130,131 might seem to superficially contradict findings that link the toxicity of the peptide as well as
limited aggregation to binding Cu. However, at neutral pH Cu2+ displaces Zn2+ from Aβ, and induces far less precipitation
than Zn2+, which might also contribute to the Aβ-reducing effects of increasing Cu levels in the cerebrospinal fluid.
Independent copper-binding sites have been identified on both Aβ and APP, which might subserve a physiological
relationship with Cu. The structure of the APP copper-binding domain has recently been solved (Protein Data Bank
accession number: 1OWT; see figure) and found to contain a novel tetrahedral copper-binding site consisting of two
histidine residues (147, 151), a tyrosine (168) and methionine (170) that favours Cu(I) coordination133. The solvent
accessibility of this site, structural homology to copper chaperones, and the role of APP in neuronal copper homeostasis are
consistent with APP acting as a neuronal metallotransporter. Intriguingly, it appears as though messenger RNA of APP is
iron regulated through the APP 5’-untranslated region and this also points to a role for iron in the metabolism of APP134.

DEMENTIA
Mental deterioration of organic
or functional origin.
SUBSTANTIA NIGRA
A small area of the brain
containing a cluster of dark-
pigmented nerve cells that
produces dopamine for
neurotransmission.
zinc and iron have been probed using nuclear magnetic
resonance and electron paramagnetic resonance spec-
troscopy55,56. Both in aqueous solution and in membrane-
mimetic environments, coordination of the metal ion is
consistent, establishing that the three histidine residues
are all involved in the coordination, along with an oxygen
ligand. Metal coordination is a highly cooperative event,
and the cooperativity is abolished by methylation of the
histidine side chains; this indicates that the second coop-
erative site would result in a histidine residue(s) bridging
between metal ions with the resulting coordination
sphere reminiscent of that observed in SOD1 (FIG. 2).
In addition to this site, Aβ is able to bind another 2.5
equivalents of metal57, and although the nature of these
other lower-affinity sites has yet to be characterized at
least one of them is redox active51 and might be impor-
tant in ROS generation.
Reactions of Cu2+ with Aβ result in the oxidative
modification of the sulphur atom of Met35 (REF. 58) and
can cause covalent crosslinking of Aβ that yields soluble
oligomers and other adducts57,59. The various forms of
Aβ in the AD brain are usually oxidatively modified60.
We have proposed that such oxidative modification
might contribute to the release of abnormally soluble
forms of Aβ from the membrane; it is these soluble forms
that correlate with DEMENTIA61. Aβ with sulphur of Met35
oxidized is more soluble and has a lower affinity for
membranes than the reduced form of Aβ, but is still
toxic to neurons58. In the normal brain, most of the Aβ
is associated with membranes62, and our structural
model of high-affinity Cu/Zn binding to Aβ indicates
that Aβ will form a membrane-embedded hexamer
when bound in that form55,56,63. We have developed the
hypothesis that when Aβ becomes hyper-metallated
by redox-active metal ions as a consequence of age-
dependent elevations in tissue metal concentrations28,
subsequent oxidation releases soluble oxidized forms of
the peptide that resist clearance. This might explain how
zinc originating from the synapse becomes so enriched
in amyloid in AD. Our model suggests that in health
soluble Aβ is not present in the cortical synapse. In AD,
soluble oxidized Aβ accumulates within the synapse, at
which the high Zn2+ concentrations precipitate the
copper/iron-metallated Aβ, creating a reservoir of
potentially toxic Aβ that is in dissociable equilibrium
with the soluble pool. The Zn2+ in the amyloid mass
partially quenches H2O2 production, which explains
why plaque amyloid burden correlates poorly with
clinical dementia54, whereas soluble Aβ levels correlate
well with clinical severity61.
Although most researchers consider Aβ to be toxic
junk, data are emerging which suggest that Aβ could
have an important physiological role as an antioxidant,
and that this function is corrupted as a result of the
aging process; evidence supporting this concept has
recently been reviewed30,64.
Parkinson’s disease. PD is characterized by the loss of
dopaminergic neurons of the SUBSTANTIA NIGRA and the
deposition of intracellular inclusion bodies. The principal

208 | MARCH 2004 | VOLUME 3 www.nature.com/reviews/drugdisc

 REVIEWS

Deposition

Aβ + M Aβ/M (Aβ/M)2 (Aβ/M)n

Excitotoxicity Memantine

MPAC

Extracellular H2O2 Calcium

dysregulation

•OH– modified Aβ ROS Intracellular H2O2

Antioxidants

Figure 2 | Oxidative stress in Alzheimer’s disease. Aβ coordinates the metal ions Zn, Fe and Cu, which induces aggregation
and, in the case of Fe and Cu, the generation of H2O2. The H2O2 can initiate a number of different events, including Fenton reactions
to form toxic hydroxyl radicals and calcium dysregulation. As calcium is pivotal in signal transduction, it can induce further
production of ROS and elicit an excitotoxicity response. Therapeutic strategies either in the clinic or undergoing clinical development
are noted in yellow boxes. Memantine, which is presently used in the clinic, inhibits excitotoxicity by targeting the N-methyl-D-
aspartate receptor. Antioxidants deactivate the generated ROS and MPACs seek to inhibit metal interactions with Aβ and prevent
the subsequent formation of ROS. MPACs, metal–protein attenuating compound; ROS, reactive oxygen species.

protein component of these deposits is α-synuclein65,
which is ubiquitously expressed in the brain; mutations
of α-synuclein (A30P and A53T) contribute to familial
forms of the disease66.
A characteristic feature of the neurons within the sub-
stantia nigra is the age-dependent accumulation of neuro-
melanin67. In PD, these neuromelanin-containing cells
are most likely to be lost68. Neuromelanin is a dark brown
pigment that accumulates metal ions, particularly iron.
Although the composition of neuromelanin has not been
rigorously characterized, it is known that it consists pri-
marily of the products of dopamine redox chemistry69,70.
Dopamine is an essential neurotransmitter, but as it is
a catechol it is also a good metal chelator, and a potential
electron donor (that is, a metal reductant). Dopamine
coordinates metals such as Cu2+ and Fe3+ (REF. 71), reduces
the oxidation state of the metal, and subsequently engen-
ders production of H2O2, setting up conditions for
Fenton chemistry. Synthetic melanins are produced by
incubating dopamine with Cu2+ and Fe3+ (REF. 72). The
purpose (if any) of neuromelanin is unknown, but it has
been postulated that it protects against dopamine-
induced redox-associated toxicity73,74. At low iron con-
centrations, melanins are known to have antioxidant
properties, but at higher metal loads melanins are pro-
oxidant75. Another postulated role for neuromelanin is as
an iron-storage molecule. Double et al.72 have shown
that neuromelanin isolated from human substantia
nigra has both high- and low-affinity Fe3+-binding sites,
and that the iron bound to neuromelanin is redox
active76. The oxidative stress associated with PD could be
the result of a breakdown in the regulation of dopamine
(neuromelanin)/iron biochemistry (FIG. 3).
A diverse array of evidence is emerging that α-synu-
clein has a role in modulating the activity of dopamine.
The A53T mutation associated with familial PD impairs
vesicular storage of dopamine77,78, which leads to the
accumulation of dopamine in the cytoplasm and subse-
quent generation of ROS through its interaction with
iron, a process that increases with age. The mutations in
α-synuclein have been shown to alter the expression of
dihydropteridine reductase, which indirectly regulates
the synthesis of dopamine79. Co-immunoprecipitation
experiments have shown that α-synuclein forms stable
complexes with the human dopamine transporter,
thereby inhibiting uptake of dopamine by its trans-
porter80 and that α-synuclein can regulate dopamine
synthesis by inhibiting tyrosine hydroxylase81. The link
between α-synuclein and redox chemistry associated
with iron-bound dopamine/neuromelanin has been
given further credence by a study showing that initiation
of Lewy body formation coincides with α-synuclein
deposition exclusively within lipofuscin and neuro-
melanin deposits82; in addition, α-synuclein crosslinked
to neuromelanin has been reported83. The breakdown in
α-synuclein-modulated dopamine homeostasis is con-
sistent with the recent observation that the pathogenicity
of mutant α-synuclein is dopamine dependent84.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 3 | MARCH 2004 | 2 0 9

REVIEWS
Aggregation
Fe/ROS
α-synuclein
Vesicular DA DA
Fe
Fe
Neuron
stimulation
MPP+ ROS NM
Fe
DA
release
Oxidative stress
DA
Figure 3 | Oxidative stress in Parkinson’s disease. Dopamine (DA) required for neuronal
signalling is vesicle bound and redox inert; when DA is released from the vesicle into the
cytoplasm it is able to coordinate Fe and undergo redox reactions that result in the formation of
neuromelanin (NM) and reactive oxygen species (ROS). Neuromelanin will also coordinate Fe
and produce ROS. The equilibrium between vesicle bound dopamine and cytoplasmic
dopamine is regulated by α-synuclein; mutations in this protein shift the dopamine equilibrium
in favour of the cytoplasm. In the presence of Fe and under conditions of oxidative stress
α-synuclein will aggregate and form deposits. MPP, 1-methyl-4-phenyl pyridine.
In addition to the regulation of dopamine by α-syn-
uclein, studies have shown a direct interaction of
α-synuclein with metal ions, leading to protein aggre-
gation85–87. Methionine oxidation inhibits α-synuclein
aggregation; however, in the presence of certain metal
ions aggregation and fibrillization of α-synuclein still
occurs88, which highlights a potential role of metals and
oxidative stress in the deposition of α-synuclein.
Amyotrophic lateral sclerosis. ALS is distinguished by the
loss of the lower motor neurons of the spinal cord and
upper motor neurons in the cerebral CORTEX; as with AD
and PD there are both sporadic and familial forms of the
disease. Like the other neurodegenerative diseases, ALS is
characterized by the deposition of a misfolded protein in
neural tissue, in this instance copper/zinc SOD89. There
are more than 100 mutations of SOD associated with the
familial forms of the disease. Through transgenic mouse
studies it has been shown that these mutations lead to a
toxic gain of function by SOD90. The nature of this gain
of function is widely debated, and there are two main
theories: one suggests that the toxicity is due to misfolded
aggregated forms of SOD, whereas the other proposes
that SOD becomes a pro-oxidant protein generating
ROS. The merits or otherwise of these two hypotheses
have recently been reviewed91, and as the aggregation
CORTEX
The unmyelinated neurons (the
mechanism lies outside the scope of this review it will
grey matter) forming the outer not be discussed here, except to note that the two theories
layer of the cerebrum. might not be mutually exclusive.
210 | MARCH 2004 | VOLUME 3
Mutations of SOD, a cupro-enzyme that detoxifies
the ROS superoxide, can convert the protein from an
anti-oxidant to a pro-oxidant capable of causing oxida-
tive insults. Evidence that inappropriate metal-mediated
redox chemistry is central to the progression of ALS
includes the observation that copper chelators inhibit
the course of the disease in both cell culture and mouse
models92–95. These initial observations suggested that the
toxicity associated with mutant SOD is the result of a
corruption of the active site of this enzyme96. However,
if the copper at the active site of SOD is the culprit
behind the toxicity, then knocking out the SOD copper
chaperone (CCS), which loads copper into the active
site, should abolish the observed toxicity. To test this
hypothesis, Subramaniam97 crossed CCS knockout mice
with an SOD mutant ALS mouse model; the phenotype
of this cross showed reduced SOD activity, which is con-
sistent with a low copper load in the active site, but the
mice still developed the ALS phenotype. These results
indicate that ALS is not due to a corruption of the
active site. However, these experiments did not examine
the possibility that other redox-active, lower-affinity
metal-binding sites exist on SOD.
Two such sites have been identified: copper can be
incorrectly incorporated into the zinc site98, and in vitro
studies99 with an H46R mutant SOD linked to familial
ALS, and which has no SOD activity, have shown that a
surface-exposed cysteine residue in SOD is also capable
of coordinating copper and is redox active. As the con-
centration of copper rises with age28, the possibility that
these lower-affinity metal-binding sites are occupied is
increased and the hypothesis that these sites might be
responsible for aberrant redox chemistry (FIG. 4), and
the generation of ROS and subsequent toxicity,
remains untested100.
Therapeutic options
For drugs to be effective treatments of neurodegenera-
tive diseases they must be capable of penetrating the
blood–brain barrier (BBB) (BOX 3).
Antioxidants. Antioxidants are molecules that react
preferentially with ROS to inactivate them and have
excited great interest because of their therapeutic
potential. A number of studies have shown beneficial
effects of antioxidants in cell culture toxicity studies
(reviewed in REF. 101). Antioxidants are part of our regular
dietary requirements and numerous claims have been
made for a variety of antioxidant supplements, including
vitamin C, ubiquinone lipoic acid, β-carotene, creatine,
melatonin, curcumin and the red-wine micronutrients,
which have a high content of phenolic and flavonoid
compounds.
Despite the large volume of positive cell culture data,
there has been limited clinical evaluation of antioxidants;
α-tocopherol (vitamin E) has been evaluated in both AD
and PD, and was found to have no beneficial effects in
PD102. However, positive effects were noted in the AD
trial103, with an increase in the median survival time of
230 days for the treated group and a significant delay
in institutionalization, but there was no improvement in
www.nature.com/reviews/drugdisc

2H+
Cu/ZnSOD H2O2
2O2• –
Age-dependent
increase in Cu
O2
Cu/ZnSOD H2O2 + OONO–
NO
Cu
Low-affinity bound Cu ROS oxidative stress
Figure 4 | Oxidative stress in amyotrophic lateral sclerosis.
The normal function of superoxide dismutase (SOD) is to
convert toxic superoxide radicals into H2O2 that are subse-
quently inactivated by catalase. In amyotrophic lateral sclerosis
this antioxidant protein is converted into a pro-oxidant protein.
With age-dependent increases in copper levels, low-affinity
copper sites on SOD such as Cys111 are occupied; these sites
are redox active and give rise to aberrant redox chemistry and
subsequent reactive oxygen species (ROS) generation.
cognitive test scores. A further trial to determine whether
vitamin E can delay the progression of mild cognitive
impairment in AD is presently ongoing104.
One consequence of ROS generation is the initiation
of excitotoxicity; this is modulated through the over-
activation of glutamate receptors, and a number of
drugs that target these receptors have shown some effi-
cacy in treating neurodegenerative diseases. Memantine,
which targets the N-methyl-D-aspartate (NMDA)
receptor, has received FDA approval for use in AD.
Memantine slows the development of the disease and is
of modest benefit to patients in the moderately severe to
Box 3 | The blood–brain barrier
For a potential therapy targeting a neurological condition, the blood–brain barrier
(BBB) is a major hurdle that must be overcome. The BBB represents the junction at
which the bloodstream meets the neuronal environment and is relatively impermeable
to fluctuations in blood Cu, Fe and Zn levels (for example, post-prandially). A barrier
consisting of brain microvessel endothelial cells ensures that neurotoxic metabolites are
excluded from the brain, but allows essential nutrients to pass across. The prevailing
opinion is that small (<500 Da), compact hydrophobic molecules have the best chance
of crossing the BBB by passive diffusion135. A number of predictors of BBB penetration
have been developed but the two most routinely used measures for predicting the
ability of a compound to penetrate the BBB are octanol partition coefficients and polar
surface areas (PSA). Octanol/water partition coefficients (log P), either computationally
or experimentally determined, are used as a measure of lipophilicity of a compound;
a rough guide suggests that a log P of at least 1.5 is required for adequate penetration of
the BBB, and the larger log P the greater the chance of a molecule passing the BBB136.
There is clearly a practical limit to how hydrophobic a molecule can be before decreases
in general solubility make a molecule unsuitable for therapeutic application. PSA is a
measure of a molecule’s compactness, a smaller PSA being more favourable for passage
across the BBB. Molecules with a PSA of 30–40 Å2 are ideal, whereas molecules with a
PSA of greater than 70 Å2 do not readily cross the BBB137.
NATURE REVIEWS | DRUG DISCOVERY
REVIEWS
severe range of the disease105. (Recent developments
with memantine have been reviewed in REFS 106,107.)
Amantadine, a partial NMDA antagonist, has been used
in the treatment of PD for some time with modest
clinical gains, including increased survival times108.
Riluzole is the only drug presently used to treat ALS
with any proven efficacy — the effects of Riluzole are
also modest but it is able to extend mean survival time
of patients by several months109. The mode of action of
Riluzole is not fully understood but it is believed to
inhibit glutamate release (and therefore to reduce excito-
toxicity) by interfering with sodium channels.
Neurotrophins have protective effects against oxida-
tive stress110; however, therapeutic applications are limited
by a lack of BBB permeability. Although efforts are con-
tinuing with regard to the design of small-molecule
stimulators of endogenous neurotrophins, an alterna-
tive approach to increasing endogenous neurotrophins
— for example, brain-derived neurotrophic factor,
which is known to enhance learning and memory, and
also protect against oxidative stress — is caloric restric-
tion. It has been proposed that a restricted diet with
limited calorie intake induces a mild cellular stress
response, which in turn upregulates proteins designed
to promote growth and survival of neurons111.
Modulation of metal–protein interactions. The modest
effects attained with antioxidants and therapies targeting
downstream effects of ROS generation, such as those
targeting excitotoxicity, are not too surprising when one
considers that they do not target the underlying cause of
the oxidative stress, but rather the consequences. A treat-
ment strategy for oxidative stress is likely to be more
effective if it targets the origin of ROS generation. A
breakdown in metal homeostasis, corruption of metallo-
enzymes or a breakdown in endogenous antioxidant
defences are the underlying causes of oxidative stress.
A pharmacological therapy targeting these interac-
tions requires the identification of molecules that will
inhibit the deleterious effects of aberrant metal interac-
tions. One approach is to design molecules that com-
pete with the target protein for the metal ions, that is, a
metal–protein attenuating compound (MPAC). This
strategy should not be confused with the concept of
‘chelation therapy’, a term associated with the removal
of bulk metals, such as in Wilson’s disease (Cu) and
β-thallesemia (Fe). The breakdown in metal homeo-
stasis in these diseases leads to tissue saturation with
metal. The mechanism of action of traditional chela-
tors is to sequester peripherally and clear by excretion.
The intention of the MPAC is to disrupt an abnormal
metal–protein interaction to achieve a subtle reparti-
tioning of metals and a subsequent normalization of
metal distribution. Once the toxic cycle is inhibited,
endogenous clearance processes can cope more effec-
tively with the accumulated protein.
The first attempt to clinically target metals in neuro-
degenerative diseases was the use of desferrioxamine in
AD. In this trial, patients were given intramuscular
injections twice daily, five days per week, for 24 months,
a treatment regime that resulted in a significant reduction
VOLUME 3 | MARCH 2004 | 2 1 1
REVIEWS

in the rate of decline of daily living skills when com-
pared with control patients112.
In light of the interactions between Aβ and metals,
we have investigated the potential use of MPACs in treat-
ing AD. We identified a hydrophobic moderate metal
chelator (Clioquinol (CQ, 5-chloro-7-iodo-8-hydroxy-
quinoline)) that is capable of crossing the BBB as a
prototypic MPAC. CQ is able to bind a range of metal
ions with moderate affinity, that is, in the nM range.
CQ was given orally in a blinded study to Tg2576
transgenic mice113. The results showed a 49% decrease in
brain Aβ burden compared with non-treated controls
after nine weeks, with no evidence of toxicity; in addi-
tion, the general health and body weight parameters
were more stable in the treated animals. Treatment with
CQ did not lead to a systematic decrease in metal levels,
which is most probably due to the drug’s moderate
binding affinities. Interestingly, in the same study, TETA,
a hydrophilic high-affinity metal chelator that is inca-
pable of crossing the BBB and which has been used to
treat Wilson’s disease, did not inhibit Aβ deposition,
indicating that systemic depletion of metal-ions (chela-
tion therapy) is unlikely to be an effective therapeutic
strategy for the treatment of AD.
The success of CQ in the transgenic mouse trials
encouraged the testing of this molecule in clinical
trials. The effects of oral CQ treatment in a random-
ized, double-blind, placebo-controlled pilot Phase II
clinical trial of moderately severe AD patients were
evaluated114. The effect of treatment was statistically
significant in preventing cognitive deterioration during
a 36-week period in the more severely affected patients
(baseline Alzheimer’s Disease Assessment Scale —
cognitive subscale ≥ 25). There was also a significant
decline in plasma Aβ42 in the CQ group compared with
an increase in the placebo group.
The success of CQ as a potential approach for AD
encouraged the testing of this molecule in animal
models of PD. The 1-methyl-4-phenyl-1,2,3,6-
tetrapyridine (MPTP) model of PD is a toxicity para-
digm, in which PD-like symptoms are induced by a
toxin that targets the dopaminergic pathways (FIG. 3),
and cells with high neuromelanin content are more
susceptible to the toxin115. The toxicity induced by
MPTP was inhibited by CQ and by overexpression of
the iron-binding protein ferritin116. These results are
consistent with the notion that 8-hydroxyquinolines
are higher-affinity ligands for iron than dopamine71.
Similar results have been reported in the 6-hydroxy-
dopamine-induced lesion model of PD that was treated
with an unspecified lipophilic derivative of 1,2-bis-
(2-amino-phenoxy)ethane-N,N,N’,N’-tetraacetic
acid, although no details of these experiments have
been published117.
Summary
The brain — an organ that requires high metal ion
concentrations to maintain many of its functions —
has a poor capacity to cope with oxidative stress, and
demonstrates little regenerative capacity. Data are
emerging that implicate aberrant redox metal interac-
tions with key proteins in the neurodegenerative dis-
eases AD, PD and ALS, and the subsequent induction of
oxidative stress leading to neurodegeneration. Oxidative
stress manifests its toxic effects through a variety of
different pathways, which provide a number of potential
therapeutic strategies. Antioxidants targeting ROS, and
MPACs targeting the upstream metal–protein interac-
tions that generate ROS, have shown promise in pre-
liminary clinical studies. Drugs that target excitotoxicity,
a downstream consequence of oxidative stress, are now
in clinical use for all three diseases.

1. Selley, M. L., Close, D. R. & Stern, S. E. The effect of
increased concentrations of homocysteine on the
concentration of (E)-4-hydroxy-2-nonenal in the plasma
and cerebrospinal fluid of patients with Alzheimer’s
disease. Neurobiol. Aging 23, 383–388 (2002).
2. Butterfield, D. A., Castegna, A., Lauderback, C. M. & Drake, J.
Evidence that amyloid β-peptide-induced lipid peroxidation
and its sequelae in Alzheimer’s disease brain contribute to
neuronal death. Neurobiol. Aging 23, 655–664 (2002).
3. Dexter, D. T. et al. Basal lipid peroxidation in substantia
nigra is increased in Parkinson’s disease. J. Neurochem.
52, 381–389 (1989).
4. Pedersen, W. A. et al. Protein modification by the lipid
peroxidation product 4-hydroxynonenal in the spinal cords
of amyotrophic lateral sclerosis patients. Ann. Neurol. 44,
819–824 (1998).
5. Arlt, S., Beisiegel, U. & Kontush, A. Lipid peroxidation in
neurodegeneration: new insights into Alzheimer’s disease.
Curr. Opin. Lipidol. 13, 289–294 (2002).
6. Sayre, L. M., Smith, M. A. & Perry, G. Chemistry and
biochemistry of oxidative stress in neurodegenerative
disease. Curr. Med. Chem. 8, 721–738 (2001).
7. Gabbita, S. P., Lovell, M. A. & Markesbery, W. R. Increased
nuclear DNA oxidation in the brain in Alzheimer’s disease.
J. Neurochem. 71, 2034–2040 (1998).
8. Alam, Z. I. et al. Oxidative DNA damage in the parkinsonian
brain: an apparent selective increase in 8-hydroxyguanine
levels in substantia nigra. J. Neurochem. 69, 1196–1203
(1997).
9. Zemlan, F. P., Thienhaus, O. J. & Bosmann, H. B.
Superoxide dismutase activity in Alzheimer’s disease:
possible mechanism for paired helical filament formation.
Brain Res. 476, 160–162 (1989).
10. Pappolla, M. A., Omar, R. A., Kim, K. S. & Robakis, N. K.
Immunohistochemical evidence of oxidative stress in
Alzheimer’s disease. Am. J. Pathol. 140, 621–628 (1992).
11. Ferrari, C. K. B. Free radicals, lipid peroxidation and
antioxidents in apoptosis: implications in cancer,
cardiovascular and neurological diseases. Biologia 55,
581–590 (2000).
12. Lovell, M. A., Xie, C. & Markesbery, W. R. Acrolein, a product
of lipid peroxidation, inhibits glucose and glutamate uptake
in primary neuronal cultures. Free Radic. Biol. Med. 29,
714–720 (2000).
13. Keller, J. N. et al. Impairment of glucose and glutamate
transport and induction of mitochondrial oxidative stress
and dysfunction in synaptosomes by amyloid β-peptide:
role of the lipid peroxidation product 4-hydroxynonenal.
J. Neurochem. 69, 273–284 (1997).
14. Mark, R. J., Hensley, K., Butterfield, D. A. & Mattson, M. P.
Amyloid β-peptide impairs ion-motive ATPase activities:
evidence for a role in loss of neuronal Ca2+ homeostasis and
cell death. J. Neurosci. 15, 6239–6249 (1995).
15. Tamagno, E. et al. H2O2 and 4-hydroxynonenal mediate
amyloid β-induced neuronal apoptosis by activating JNKs
and p38MAPK. Exp. Neurol. 180, 144–155 (2003).
16. Ermak, G. & Davies, K. J. Calcium and oxidative stress:
from cell signaling to cell death. Mol. Immunol. 38, 713–721
(2002).
17. LaFerla, F. M. Calcium dyshomeostasis and intracellular
signalling in Alzheimer’s disease. Nature Rev. Neurosci. 3,
862–872 (2002).
18. Gibson, G. E. Interactions of oxidative stress with cellular
calcium dynamics and glucose metabolism in
Alzheimer’s disease. Free Radic. Biol. Med. 32,
1061–1070 (2002).
19. Mattson, M. P. & Chan, S. L. Neuronal and glial calcium
signaling in Alzheimer’s disease. Cell Calcium 34, 385–397
(2003).
20. Yamamoto, K. et al. The hydroxyl radical scavenger
Nicaraven inhibits glutamate release after spinal injury in
rats. Neuroreport 9, 1655–1659 (1998).
21. Mattson, M. P. Excitotoxic and excitoprotective
mechanisms: abundant targets for the prevention and
treatment of neurodegenerative disorders. Neuromolecular
Med. 3, 65–94 (2003).
22. Lewen, A., Matz, P. & Chan, P. H. Free radical pathways in
CNS injury. J. Neurotrauma 17, 871–890 (2000).
23. Suzuki, Y. J., Forman, H. J. & Sevanian, A. Oxidants as
stimulators of signal transduction. Free Radic. Biol. Med.
22, 269–285 (1997).
A seminal work describing a potential physiological
role for ROS.
24. Neill, S., Desikan, R. & Hancock, J. Hydrogen peroxide
signalling. Curr. Opin. Plant Biol. 5, 388–395 (2002).
25. Halliwell, B. & Gutteridge, J. Free Radicals in Biology and
Medicine (Oxford Univ. Press, Oxford, 1999).
The definitive textbook on ROS.
26. Morita, A., Kimura, M. & Itokawa, Y. The effect of aging on
the mineral status of female mice. Biol. Trace Elem. Res. 42,
165–177 (1994).
27. Takahashi, S. et al. Age-related changes in the
concentrations of major and trace elements in the brain of
rats and mice. Biol. Trace Elem. Res. 80, 145–158 (2001).
28. Maynard, C. J. et al. Overexpression of Alzheimer’s disease
amyloid-β opposes the age-dependent elevations of brain
copper and iron. J. Biol. Chem. 277, 44670–4476 (2002).
References 26–28 describe the rise in metal
concentration within the brain as a consequence of

212 | MARCH 2004 | VOLUME 3 www.nature.com/reviews/drugdisc


normal aging, age being the major risk factor for the
neurodegenerative diseases.
29. Bush, A. I. Metals and neuroscience. Curr. Opin. Chem. Biol.
4, 184–191 (2000).
30. Bush, A. I. The metallobiology of Alzheimer’s disease.
Trends Neurosci. 26, 207–214 (2003).
31. Hardy, J. Amyloid, the presenilins and Alzheimer’s disease.
Trends Neurosci. 20, 154–159 (1997).
32. Price, D. L., Tanzi, R. E., Borchelt, D. R. & Sisodia, S. S.
Alzheimer’s disease: genetic studies and transgenic models.
Annu. Rev. Genet. 32, 461–493 (1998).
33. Masters, C. L. et al. Amyloid plaque core protein in
Alzheimer disease and Down syndrome. Proc. Natl Acad.
Sci. USA 82, 4245–4249 (1985).
34. Glenner, G. G. & Wong, C. W. Alzheimer’s disease: initial
report of the purification and characterization of a novel
cerebrovascular amyloid protein. Biochem. Biophys. Res.
Commun. 120, 885–890 (1984).
References 33 and 34 represent the first characteri-
zation of the deposited amyloid within the AD brain.
35. Kang, J. et al. The precursor of Alzheimer’s disease amyloid
A4 protein resembles a cell-surface receptor. Nature 325,
733–736 (1987).
36. Lee, J. Y., Cole, T. B., Palmiter, R. D., Suh, S. W. & Koh, J. Y.
Contribution by synaptic zinc to the gender-disparate
plaque formation in human Swedish mutant APP transgenic
mice. Proc. Natl Acad. Sci. USA 99, 7705–7710 (2002).
This paper highlights the pivotal role zinc plays in the
deposition of Aβ.
37. Lovell, M. A., Robertson, J. D., Teesdale, W. J.,
Campbell, J. L. & Markesbery, W. R. Copper, iron and zinc
in Alzheimer’s disease senile plaques. J. Neurol. Sci. 158,
47–52 (1998).
This paper illustrates that plaques associated with AD
are metal sinks.
38. Smith, M. A., Harris, P. L., Sayre, L. M. & Perry, G. Iron
accumulation in Alzheimer disease is a source of redox-
generated free radicals. Proc. Natl Acad. Sci. USA 94,
9866–9868 (1997).
39. Huang, X. et al. Zinc-induced Alzheimer’s Aβ 1–40
aggregation is mediated by conformational factors. J. Biol.
Chem. 272, 26464–26470 (1997).
40. Atwood, C. S. et al. Dramatic aggregation of Alzheimer Aβ
by Cu(II) is induced by conditions representing physiological
acidosis. J. Biol. Chem. 273, 12817–12826 (1998).
41. Terry, R. D. The pathogenesis of Alzheimer disease: an
alternative to the amyloid hypothesis. J. Neuropathol. Exp.
Neurol. 55, 1023–1025 (1996).
42. Martins, R. N., Harper, C. G., Stokes, G. B. & Masters, C. L.
Increased cerebral glucose-6-phosphate dehydrogenase
activity in Alzheimer’s disease may reflect oxidative stress.
J. Neurochem. 46, 1042–1045 (1986).
43. Sayre, L. M. et al. In situ oxidative catalysis by neurofibrillary
tangles and senile plaques in Alzheimer’s disease: a central
role for bound transition metals. J. Neurochem. 74,
270–279 (2000).
44. Atwood, C. S., Huang, X., Moir, R. D., Tanzi, R. E. &
Bush, A. I. Role of free radicals and metal ions in the
pathogenesis of Alzheimer’s disease. Met. Ions Biol. Syst.
36, 309–364 (1999).
45. Basun, H., Forssell, L. G., Wetterberg, L. & Winblad, B.
Metals and trace elements in plasma and cerebrospinal fluid
in normal aging and Alzheimer’s disease. J. Neural Transm.
Park. Dis. Dement. Sect. 3, 231–258 (1991).
46. Squitti, R. et al. Elevation of serum copper levels in
Alzheimer’s disease. Neurology 59, 1153–1161 (2002).
47. Bishop, G. M. et al. Iron: a pathological mediator of
Alzheimer disease? Dev. Neurosci. 24, 184–187 (2002).
48. Smith, M. A. et al. Cytochemical demonstration of oxidative
damage in Alzheimer disease by immunochemical
enhancement of the carbonyl reaction with 2,4-
dinitrophenylhydrazine. J. Histochem. Cytochem. 46,
731–735 (1998).
49. Dong, J. et al. Metal binding and oxidation of amyloid-β
within isolated senile plaque cores: Raman microscopic
evidence. Biochemistry 42, 2768–2773 (2003).
50. Behl, C., Davis, J. B., Lesley, R. & Schubert, D. Hydrogen
peroxide mediates amyloid β protein toxicity. Cell 77,
817–827 (1994).
Initial identification of H2O2 as a key participant in the
toxicity of Aβ.
51. Opazo, C. et al. Metalloenzyme-like activity of Alzheimer’s
disease β-amyloid. Cu-dependent catalytic conversion of
dopamine, cholesterol, and biological reducing agents to
neurotoxic H2O2. J. Biol. Chem. 277, 40302–40308
(2002).
52. Huang, X. et al. Cu(II) potentiation of alzheimer Aβ
neurotoxicity. Correlation with cell-free hydrogen peroxide
production and metal reduction. J. Biol. Chem. 274,
37111–37116 (1999).
NATURE REVIEWS | DRUG DISCOVERY
53. Huang, X. et al. The Aβ peptide of Alzheimer’s disease
directly produces hydrogen peroxide through metal ion
reduction. Biochemistry 38, 7609–7616 (1999).
References 52 and 53 demonstrated that Aβ will
generate H2O2 when Aβ coordinates copper.
54. Cuajungco, M. P. et al. Evidence that the β-amyloid plaques
of Alzheimer’s disease represent the redox-silencing and
entombment of Aβ by zinc. J. Biol. Chem. 275,
19439–19442 (2000).
55. Curtain, C. C. et al. Alzheimer’s disease amyloid-β binds
copper and zinc to generate an allosterically ordered
membrane-penetrating structure containing superoxide
dismutase-like subunits. J. Biol. Chem. 276, 20466–20473
(2001).
This paper shows that metals can affect Aβ
membrane-bound structures with a metal
coordination site similar to the active site of SOD.
56. Curtain, C. C. et al. Metal ions, pH, and cholesterol regulate
the interactions of Alzheimer’s disease amyloid-β peptide
with membrane lipid. J. Biol. Chem. 278, 2977–2982 (2003).
57. Atwood, C. S. et al. Characterization of copper interactions
with alzheimer amyloid β peptides: identification of an
attomolar-affinity copper binding site on amyloid β 1–42.
J. Neurochem. 75, 1219–1233 (2000).
58. Barnham, K. J. et al. Neurotoxic, redox-competent
Alzheimer’s β-amyloid is released from lipid membrane by
methionine oxidation. J. Biol. Chem. 278, 42959–42965
(2003).
A demonstration that a non-fibrillar, highly soluble
form of Aβ is still neurotoxic.
59. Atwood, C. S. et al. Copper catalyzed oxidation of Alzheimer
Aβ. Cell. Mol. Biol. (Noisy-Le-Grand) 46, 777–783 (2000).
60. Head, E. et al. Oxidation of Aβ and plaque biogenesis in
Alzheimer’s disease and Down syndrome. Neurobiol. Dis.
8, 792–806 (2001).
61. McLean, C. A. et al. Soluble pool of Aβ amyloid as a
determinant of severity of neurodegeneration in Alzheimer’s
disease. Ann. Neurol. 46, 860–866 (1999).
A demonstration that AD progression correlates
with soluble Aβ, not the deposited material found
within plaques.
62. Cherny, R. A. et al. Aqueous dissolution of Alzheimer’s disease
Aβ amyloid deposits by biometal depletion. J. Biol. Chem.
274, 23223–23228 (1999).
63. Lau, T. L., Barnham, K. J., Curtain, C. C., Masters, C. L. &
Separovic, F. Magnetic resonance studies of β-amyloid
peptides. Aust. J. Chem 56, 349–356 (2003).
64. Atwood, C. S. et al. Amyloid-β: a chameleon walking in two
worlds: a review of the trophic and toxic properties of
amyloid-β. Brain Res. Brain Res. Rev. 43, 1–16 (2003).
65. Spillantini, M. G. et al. α-synuclein in Lewy bodies. Nature
388, 839–840 (1997).
Identifies α-synuclein as the major protein deposited
in the Lewy bodies associated with PD.
66. Gasser, T. Genetics of Parkinson’s disease. J. Neurol. 248,
833–840 (2001).
67. Marsden, C. D. Neuromelanin and Parkinson’s disease.
J. Neural. Transm. Suppl. 19, 121–141 (1983).
68. Hirsch, E., Graybiel, A. M. & Agid, Y. A. Melanized
dopaminergic neurons are differentially susceptible to
degeneration in Parkinson’s disease. Nature 334,
345–348 (1988).
Identifies the selective destruction of specific cell
types in the progression of PD.
69. Wakamatsu, K., Fujikawa, K., Zucca, F. A., Zecca, L. & Ito, S.
The structure of neuromelanin as studied by chemical
degradative methods. J. Neurochem. 86, 1015–1023
(2003).
Identifies the chemical nature of neuromelanin
thereby providing further evidence as to why
neuromelanin-containing cells are susceptible in PD.
70. Zecca, L. et al. The neuromelanin of human substantia nigra:
structure, synthesis and molecular behaviour. J. Neural.
Transm. Suppl. 145–155 (2003).
71. Gerard, C., Chehhal, H. & Hugel, R. P. Complexes of
iron(III) with ligands of biological interest: dopamine and
8-hydroxyquinoline-5-sulfonic acid. Polyhedron 13,
591–597 (1994).
72. Double, K. L. et al. Iron-binding characteristics of
neuromelanin of the human substantia nigra. Biochem.
Pharmacol. 66, 489–494 (2003).
73. Smythies, J. On the function of neuromelanin. Proc. R. Soc.
Lond. B Biol. Sci. 263, 487–489 (1996).
74. Sulzer, D. et al. Neuromelanin biosynthesis is driven by
excess cytosolic catecholamines not accumulated by
synaptic vesicles. Proc. Natl Acad. Sci. USA 97,
11869–11874 (2000).
75. Ben-Shachar, D., Riederer, P. & Youdim, M. B. Iron-melanin
interaction and lipid peroxidation: implications for
Parkinson’s disease. J. Neurochem. 57, 1609–1614 (1991).
REVIEWS
76. Faucheux, B. A. et al. Neuromelanin associated redox-active
iron is increased in the substantia nigra of patients with
Parkinson’s disease. J. Neurochem. 86, 1142–1148 (2003).
77. Lotharius, J. et al. Effect of mutant α-synuclein on
dopamine homeostasis in a new human mesencephalic cell
line. J. Biol. Chem. 277, 38884–38894 (2002).
78. Lotharius, J. & Brundin, P. Impaired dopamine storage
resulting from α-synuclein mutations may contribute to the
pathogenesis of Parkinson’s disease. Hum. Mol. Genet. 11,
2395–2407 (2002).
79. Baptista, M. J. et al. Co-ordinate transcriptional regulation of
dopamine synthesis genes by α-synuclein in human
neuroblastoma cell lines. J. Neurochem. 85, 957–968 (2003).
80. Wersinger, C., Prou, D., Vernier, P. & Sidhu, A. Modulation of
dopamine transporter function by α-synuclein is altered by
impairment of cell adhesion and by induction of oxidative
stress. FASEB. J. (2003).
81. Perez, R. G. et al. A role for α-synuclein in the regulation of
dopamine biosynthesis. J. Neurosci. 22, 3090–3099 (2002).
82. Braak, E. et al. α-synuclein immunopositive Parkinson’s
disease-related inclusion bodies in lower brain stem nuclei.
Acta Neuropathol. (Berl) 101, 195–201 (2001).
83. Fasano, M., Giraudo, S., Coha, S., Bergamasco, B. &
Lopiano, L. Residual substantia nigra neuromelanin in
Parkinson’s disease is cross-linked to α-synuclein.
Neurochem. Int. 42, 603–606 (2003).
84. Xu, J. et al. Dopamine-dependent neurotoxicity of α-
synuclein: a mechanism for selective neurodegeneration in
Parkinson disease. Nature Med. 8, 600–606 (2002).
References 77–84 show links between α-synuclein,
which genetics demonstrates is important in the
progression of PD, and dopamine, which is required
for the redox chemistry that is responsible for the
oxidative stress.
85. Paik, S. R., Shin, H. J., Lee, J. H., Chang, C. S. & Kim, J.
Copper(II)-induced self-oligomerization of α-synuclein.
Biochem. J. 340, 821–828 (1999).
86. Uversky, V. N., Li, J. & Fink, A. L. Metal-triggered structural
transformations, aggregation, and fibrillation of human
α-synuclein. A possible molecular NK between Parkinson’s
disease and heavy metal exposure. J. Biol. Chem. 276,
44284–44296 (2001).
87. Lee, E. N., Lee, S. Y., Lee, D., Kim, J. & Paik, S. R. Lipid
interaction of α-synuclein during the metal-catalyzed
oxidation in the presence of Cu2+ and H2O2. J. Neurochem.
84, 1128–1142 (2003).
88. Yamin, G., Glaser, C. B., Uversky, V. N. & Fink, A. L. Certain
metals trigger fibrillation of methionine-oxidized α-synuclein.
J. Biol. Chem. 278, 27630–27635 (2003).
89. Bruijn, L. I. et al. Aggregation and motor neuron toxicity of
an ALS-linked SOD1 mutant independent from wild-type
SOD1. Science 281, 1851–1854 (1998).
90. Gurney, M. E. et al. Motor neuron degeneration in mice that
express a human Cu,Zn superoxide dismutase mutation.
Science 264, 1772–1775 (1994).
This paper reports that ALS mutant SOD is toxic —
that is, there is a toxic gain of function.
91. Valentine, J. S. & Hart, P. J. Misfolded CuZnSOD and
amyotrophic lateral sclerosis. Proc. Natl Acad. Sci. USA
100, 3617–3622 (2003).
92. Wiedau-Pazos, M. et al. Altered reactivity of superoxide
dismutase in familial amyotrophic lateral sclerosis. Science
271, 515–518 (1996).
93. Ghadge, G. D. et al. Mutant superoxide dismutase-1-linked
familial amyotrophic lateral sclerosis: molecular mechanisms
of neuronal death and protection. J. Neurosci. 17,
8756–8766 (1997).
94. Hottinger, A. F., Fine, E. G., Gurney, M. E., Zurn, A. D. &
Aebischer, P. The copper chelator D-penicillamine delays
onset of disease and extends survival in a transgenic
mouse model of familial amyotrophic lateral sclerosis.
Eur. J. Neurosci. 9, 1548–1551 (1997).
95. Azzouz, M. et al. Prevention of mutant SOD1 motoneuron
degeneration by copper chelators in vitro. J. Neurobiol. 42,
49–55 (2000).
References 92–95 show that copper chelators inhibit
the mutant SOD toxicity, implying that copper has a
pivotal role in the progression of ALS.
96. Estevez, A. G. et al. Induction of nitric oxide-dependent
apoptosis in motor neurons by zinc-deficient superoxide
dismutase. Science 286, 2498–500 (1999).
97. Subramaniam, J. R. et al. Mutant SOD1 causes motor
neuron disease independent of copper chaperone-mediated
copper loading. Nature Neurosci. 5, 301–307 (2002).
Reports that the copper at the active site of SOD is
not responsible for the observed toxicity.
98. Goto, J. J. et al. Loss of in vitro metal ion binding specificity
in mutant copper-zinc superoxide dismutases associated
with familial amyotrophic lateral sclerosis. J. Biol. Chem.
275, 1007–1014 (2000).
VOLUME 3 | MARCH 2004 | 2 1 3
REVIEWS
99. Liu, H. et al. Copper2+ binding to the surface residue cysteine
111 of His46Arg human copper-zinc superoxide dismutase,
a familial amyotrophic lateral sclerosis mutant. Biochemistry
39, 8125–8132 (2000).
Identification of a low-affinity, surface-located
copper-binding site on SOD that is redox active and
therefore could play some role in mediating redox-
associated oxidative stress.
100. Bush, A. I. Is ALS caused by an altered oxidative activity of
mutant superoxide dismutase? Nature Neurosci. 5, 919;
author reply 919–920 (2002).
101. Moosmann, M. & Behl, C. Antioxidants as treatment for
neurodegenerative disorders. Exp. Opin. Invest. Drugs 11,
1407–1435 (2002).
102. Effects of tocopherol and deprenyl on the progression of
disability in early Parkinson’s disease. The Parkinson Study
Group. N. Engl. J. Med. 328, 176–183 (1993).
103. Sano, M. et al. A controlled trial of selegiline, α-tocopherol,
or both as treatment for Alzheimer’s disease. The
Alzheimer’s Disease Cooperative Study. N. Engl. J. Med.
336, 1216–1222 (1997).
This study showed that the antioxident vitamin E has
protective effects in an AD clinical trial.
104. Grundman, M. Vitamin E and Alzheimer’s disease: the basis
for additional clinical trials. Am. J. Clin. Nutr. 71,
S630–S636 (2000).
105. Reisberg, B. et al. Memantine in moderate-to-severe Alz-
heimer’s disease. N. Engl. J. Med. 348, 1333–1341 (2003).
106. Mobius, H. J. Memantine: update on the current evidence.
Int. J. Geriatr. Psychiatry 18, S47–S54 (2003).
107. Winblad, B. & Jelic, V. Treating the full spectrum of dementia
with memantine. Int. J. Geriatr. Psychiatry 18, S41–S46 (2003).
108. Hely, M. A., Fung, V. S. & Morris, J. G. Treatment of
Parkinson’s disease. J. Clin. Neurosci. 7, 484–494 (2000).
109. Bensimon, G., Lacomblez, L. & Meininger, V. A controlled
trial of riluzole in amyotrophic lateral sclerosis. ALS/Riluzole
Study Group. N. Engl. J. Med. 330, 585–591 (1994).
110. Doraiswamy, P. M. Non-cholinergic strategies for treating
and preventing Alzheimer’s disease. CNS Drugs 16,
811–824 (2002).
111. Mattson, M. P. Will caloric restriction and folate protect
against AD and PD? Neurology 60, 690–695 (2003).
112. Crapper McLachlan, D. R. et al. Intramuscular
desferrioxamine in patients with Alzheimer’s disease.
Lancet 337, 1304–1308 (1991).
113. Cherny, R. A. et al. Treatment with a copper-zinc chelator
markedly and rapidly inhibits β-amyloid accumulation in
Alzheimer’s disease transgenic mice. Neuron 30, 665–676
(2001).
Inhibiting metal Aβ interactions have positive
outcomes in a transgenic mouse model.
114. Ritchie, C. W. et al. Metal–protein attenuation with
iodochlorhydroxyquin (clioquinol) targeting Aβ amyloid
deposition and toxicity in Alzheimer’s disease: biochemical
and clinical responses in a pilot phase 2 clinical trial. Arch.
Neurol. 60, 1685–1691 (2003).
214 | MARCH 2004 | VOLUME 3
Proof-of-concept Phase II clinical trial that
demonstrates that inhibiting Aβ metal interactions has
positive clinical results.
115. Dauer, W. & Przedborski, S. Parkinson’s disease:
mechanisms and models. Neuron 39, 889–909 (2003).
116. Kaur, D. et al. Genetic or pharmacological iron chelation
prevents MPTP-induced neurotoxicity in vivo: a novel
therapy for Parkinson’s disease. Neuron 37, 899–909
(2003).
A demonstration that targeting metals in a mouse
model of PD has positive effects.
117. Angel, I. et al. Metal ion chelation in neurodegenerative
disorders. Drug Dev. Res. 56, 300–309 (2002).
118. Campuzano, V. et al. Friedreich’s ataxia: autosomal
recessive disease caused by an intronic GAA triplet repeat
expansion. Science 271, 1423–1427 (1996).
119. Bradley, J. L. et al. Clinical, biochemical and molecular
genetic correlations in Friedreich’s ataxia. Hum. Mol. Genet.
9, 275–282 (2000).
120. Rustin, P. et al. Effect of idebenone on cardiomyopathy in
Friedreich’s ataxia: a preliminary study. Lancet 354,
477–479 (1999).
121. Hausse, A. O. et al. Idebenone and reduced cardiac
hypertrophy in Friedreich’s ataxia. Heart 87, 346–349
(2002).
122. Buyse, G. et al. Idebenone treatment in Friedreich’s ataxia:
neurological, cardiac, and biochemical monitoring.
Neurology 60, 1679–1681 (2003).
123. Wong, A. et al. The Friedreich’s ataxia mutation confers
cellular sensitivity to oxidant stress which is rescued by
chelators of iron and calcium and inhibitors of apoptosis.
Hum. Mol. Genet. 8, 425–430 (1999).
124. Richardson, D. R., Mouralian, C., Ponka, P. & Becker, E.
Development of potential iron chelators for the treatment
of Friedreich’s ataxia: ligands that mobilize mitochondrial
iron. Biochim. Biophys. Acta 1536, 133–40 (2001).
125. Culotta, V. C. et al. The copper chaperone for superoxide
dismutase. J. Biol. Chem. 272, 23469–23472 (1997).
Identification of the protein responsible for loading
copper into the active site of SOD.
126. Waggoner, D. J., Bartnikas, T. B. & Gitlin, J. D. The role of
copper in neurodegenerative disease. Neurobiol. Dis. 6,
221–230 (1999).
127. Rae, T. D., Schmidt, P. J., Pufahl, R. A., Culotta, V. C. &
O’Halloran, T. V. Undetectable intracellular free copper:
the requirement of a copper chaperone for superoxide
dismutase. Science 284, 805–808 (1999).
This paper illustrates the elaborate mechanisms that
nature has evolved to tightly regulate metal ions.
128. Andrews, N. C. Mining copper transport genes. Proc. Natl
Acad. Sci. USA 98, 6543–6545 (2001).
129. White, A. R. et al. Copper levels are increased in the cerebral
cortex and liver of APP and APLP2 knockout mice. Brain
Res. 842, 439–444 (1999).
A demonstration that APP has a role in copper
homeostasis.
130. Bayer, T. A. et al. Dietary Cu stabilizes brain superoxide
dismutase 1 activity and reduces amyloid Aβ production in
APP23 transgenic mice. Proc. Natl Acad. Sci. USA 100,
14187–14192 (2003).
131. Phinney, A. L. et al. In vivo reduction of amyloid-β by a
mutant copper transporter. Proc. Natl Acad. Sci. USA
100, 14193–14198 (2003).
132. Borchardt, T. et al. Copper inhibits β-amyloid production
and stimulates the non-amyloidogenic pathway of
amyloid-precursor-protein secretion. Biochem. J. 344 Pt
2, 461–467 (1999).
References 130–132 Illustrate how copper can
mediate APP processing to reduce Aβ production.
133. Barnham, K. J. et al. Structure of the Alzheimer’s disease
amyloid precursor protein copper binding domain.
A regulator of neuronal copper homeostasis. J. Biol.
Chem. 278, 17401–17407 (2003).
3D Structure of the copper–binding domain of APP
showing where and how APP coordinates copper.
134. Rogers, J. T. et al. An iron-responsive element type II in
the 5’-untranslated region of the Alzheimer’s amyloid
precursor protein transcript. J. Biol. Chem. 277,
45518–45528 (2002).
135. Levin, V. A. Relationship of octanol/water partition
coefficient and molecular weight to rat brain capillary
permeability. J. Med. Chem. 23, 682–684 (1980).
136. Habgood, M. D. et al. Investigation into the correlation
between the structure of hydroxypyridinones and
blood–brain barrier permeability. Biochem. Pharmacol.
57, 1305–1310 (1999).
137. Iyer, M., Mishru, R., Han, Y. & Hopfinger, A. J. Predicting
blood–brain barrier partitioning of organic molecules using
membrane–interaction QSAR analysis. Pharm Res 19,
1611–1621 (2002).
Competing interests statement
The authors declare competing financial interests: see Web version
for details.
Online links
DATABASES
The following terms in this article are linked online to:
LocusLink: http://www.ncbi.nlm.nih.gov/LocusLink/
β-actin | catalase | creatine kinase | frataxin | glucose transporter
type-3 | glutathione peroxidase | glutathione reductase |
mitogen-activated protein kinase-1 | SOD | α-synuclein |
xanthine dehydrogenase
Online Mendelian Inheritance in Man:
http://www.ncbi.nlm.nih.gov/Omim/
Alzheimer’s disease | amyotrophic lateral sclerosis |
Friedreich’s ataxia | Huntington’s disease | nitric oxide synthase |
Parkinson’s disease | stroke | Wilson’s disease
Access to this interactive links box is free online.
www.nature.com/reviews/drugdisc