J Mol Med (2015) 93:481–487

DOI 10.1007/s00109-015-1273-3

REVIEW

Metal-deficient SOD1 in amyotrophic lateral sclerosis

James B. Hilton & Anthony R. White & Peter J. Crouch

Received: 17 December 2014 / Revised: 24 February 2015 / Accepted: 25 February 2015 / Published online: 11 March 2015
# The Author(s) 2015. This article is published with open access at Springerlink.com

Abstract Mutations to the ubiquitous antioxidant en- Introduction

zyme Cu/Zn superoxide dismutase (SOD1) were the
first established genetic cause of the fatal, adult-onset Amyotrophic lateral sclerosis (ALS) is a fatal neurode-
neurodegenerative disease amyotrophic lateral sclerosis generative disease that selectively afflicts the motor neu-
(ALS). It is widely accepted that these mutations do rons in the spinal cord, motor cortex and brainstem
not cause ALS via a loss of antioxidant function, but leading to their dysfunction and eventual death [1].
elucidating the alternate toxic gain of function has prov- The loss of functional motor neurons and associated
en to be elusive. Under physiological conditions, SOD1 clinical symptoms in ALS is progressive, and despite
binds one copper ion and one zinc ion per monomer to variability in timeframes relative to clinical subtype,
form a highly stable and functional homodimer, but ALS ultimately causes paralysis and premature death
there is now ample evidence to indicate aberrant persis- which is generally due to respiratory failure. In general,
tence of SOD1 in an intermediate metal-deficient state the peak age of onset is 45–60 years, and survival post
may contribute to the protein’s involvement in ALS. diagnosis in the vicinity of 3–5 years.
This review briefly discusses some of the data to sup- The predominant proportion of ALS cases is
port a role for metal-deficient SOD1 in the development characterised as sporadic, with only 5–10 % of cases
of ALS and some of the outcomes from drug develop- presenting with a known heritable genetic basis [2].
ment studies that have aimed to modify the symptoms The most prominent and well-understood genetic basis
of ALS by targeting the metal state of SOD1. The implica- for familial ALS concerns point mutations to the ubiq-
tions for the metal state of SOD1 in cases of sporadic ALS that uitously expressed antioxidant enzyme superoxide dis-
do not involve mutant SOD1 are also discussed. mutase 1 (SOD1) [3]. SOD1 mutations were the first
described genetic cause of ALS [4] and are accordingly
the most widely studied. They may present as missense
Keywords Amyotrophic lateral sclerosis (ALS) . Motor mutations or frameshift mutations leading to protein
neuron disease (MND) . Copper (Cu) . Zinc (Zn) . Cu/Zn truncation and can reside throughout the protein includ-
superoxide dismutase (SOD1) . Protein misfolding . ing within key catalytic and structural regions such as
Diacetylbis(4-methylthiosemicarbazonato)copperII the metal-binding regions and dimer interface [5]. SOD1
mutations are typically segregated into two distinct
groups termed the Bwild-type-like^ mutants (G37R,
G93A, A4V, etc.), which retain similar properties to
J. B. Hilton : A. R. White : P. J. Crouch (*)
wild-type SOD1, and the metal-binding region mutants
Department of Pathology, The University of Melbourne, (G85R, H46R, etc.), characterised by vastly different
Melbourne, Victoria 3010, Australia biophysical attributes related to metal binding, SOD1
e-mail: [email protected] enzymatic activity and spectroscopy [6–9]. But despite
A. R. White : P. J. Crouch
gaining good research attention over the past two de-
Florey Institute of Neuroscience and Mental Health, The University cades, the definitive mechanisms by which SOD1 muta-
of Melbourne, Melbourne, Victoria 3010, Australia tions cause ALS are yet to be fully elucidated.
482 J Mol Med (2015) 93:481–487

Mutant SOD1 rodent models of ALS [18, 19] led to the postulation that the formation of these
aberrant inclusions may represent the toxic gain-of-function
As SOD1 plays a vital role as an antioxidant converting su- mechanism through which SOD1-induced motor neuron
peroxide (O2.−) into hydrogen peroxide (H2O2) for subsequent death occurs. The production of metal-deficient monomeric
breakdown by glutathione peroxidase and catalase into water species destabilised in a disulphide bond reduced state aided
[10], it was initially suspected that a loss of function may through mutations could present a pathway through which
contribute to ALS pathogenesis. This hypothesis was subse- insoluble aggregates can form [20, 21]. Ubiquitin detected
quently rejected since SOD1 knockout mice failed to develop within the aggregates [22] suggested that these insoluble in-
an ALS-like phenotype [11]. These mice did display an in- clusions were marked for proteasomal degradation which may
creased motor neuron sensitivity to axonal injury [11] and have been impeded, perhaps through overloading of the
pathology in response to paraquat induced of oxidative stress ubiquitin-proteasome system [23]. Moreover, the detection
[12], demonstrating the clear requirement for SOD1 in of other proteins such as chaperones within the aggregates
protecting motor neurons from adverse insults, but they also [24] indicated the possibility of a secondary loss-of-function
provided early evidence that a loss of SOD1 function in iso- disease mechanism through which various proteins become
lation is not a sole contributor to ALS in mutant SOD1 cases aberrantly sequestered and co-localise within the SOD1 ag-
of the disease. Thus, in the absence of an overt phenotype in gregates [25]. The fact that observable insoluble aggregates
the SOD1 knockout mice, the main contention is that a toxic were typically present at late stages in the disease progression
gain of function is fundamentally responsible for the develop- [26], however, suggested that aggregates are likely to be a
ment of ALS in cases where SOD1 mutations are involved. downstream consequence of alternative pathological process-
To more thoroughly investigate the role played by the myriad es. With the advent of monoclonal antibodies designed to bind
SOD1 mutants discovered in ALS, various rodent models of the the putative toxic misfolded species of SOD1, attention has
disease have been generated by overexpressing human SOD1 shifted towards a soluble form of misfolded SOD1 which was
containing different point mutations. The initial mutant SOD1 reported to be detectable across all ages in mutant SOD1 ex-
rodent models were developed soon after mutant SOD1 was pressing murine spinal cord tissue [27]. Supporting this find-
identified as a cause of ALS in humans, and depending on the ing, it was observed that co-expression of mutant and wild-
form and amount of mutant SOD1 expressed, the rate of disease type human SOD1 increased the solubility of mutant
progression and lifespan of these animals varies considerably misfolded SOD1 species leading to greater cellular toxicity
[13, 14]. Although the overexpression of mutant human SOD1 [28], demonstrating that the soluble form may be an earlier
in these animals is not an ideal representation of disease condi- contributor to ALS pathogenesis whilst aggregate formation
tions in ALS, due to reliance on overexpression of an exogenous may represent a cellular defence mechanism [28].
gene, these animals nonetheless provide a useful and widely
available model for disease investigation. Furthermore, the de-
velopment of transgenic rodents overexpressing the wild-type Zinc-deficient SOD1
form of human SOD1 has provided the opportunity to investigate
specific differences between mutant and wild-type SOD1 within Once translated, the incipient monomeric polypeptide of
the constraints of an overexpression model. Wild-type SOD1- SOD1 becomes bound to one zinc atom providing structural
overexpressing mice are generally used as a negative control line integrity before the direct interaction with copper chaperone
due to the absence of an overt ALS-like phenotype in the hemi- for superoxide dismutase (CCS) which ensures the binding of
zygous, wild-type SOD1-expressing mice, particularly when a copper atom required for catalytic activity [29]. The estab-
compared to age-matched mutant SOD1 expressing mice. How- lishment of a fully functional SOD1 enzyme then occurs
ever, wild-type SOD1-expressing mice have nonetheless provid- through intramolecular disulphide bond formation and com-
ed some evidence to implicate a role for SOD1 in cases of ALS bination with another monomeric protein to create the func-
that do not involve mutant SOD1; subtle phenotype changes tional homodimer [30]. The fact that many disparate point
have been reported for hemizygous, wild-type SOD1-overex- mutations within SOD1 can induce ALS led to the hypothesis
pressing mice at a relatively young age [15], whereas a more that an impaired folding mechanism may be responsible for
recent study has described pronounced ALS-like features in ho- producing aberrant copper-mediated chemistry. Changes in
mozygous, wild-type SOD1-overexpressing mice [16]. SOD1 protein folding due to mutations may allow aberrant
substrate access to the copper-dependent catalytic region lead-
ing to potentially harmful reactions. Furthermore, impairment
SOD1 aggregates in copper-binding capacity may promote its release where it
could catalyse reactions that cause oxidative damage [31]. An
The discovery of insoluble aggregates of SOD1 in mutant initial substrate proposed to induce cellular pathology was
SOD1 mouse models of ALS [17] and human cases of ALS peroxynitrite [32] which is created spontaneously through
J Mol Med (2015) 93:481–487 483

interaction between nitric oxide (NO) and O2.−, leading to death and the addition of fully metallated wild-type SOD1
tyrosine nitration of proteins capable of interfering with cellu- acting to increase toxicity in the presence of NO; it was dem-
lar pathways. However, the fact that the G85R SOD1 muta- onstrated that the fully metallated monomer of wild-type
tion causes rapidly progressing disease without retaining SOD1 could form a dimer with a zinc-deficient monomer of
copper-dependent enzymatic activity [33, 34], in conjunction mutant SOD1, thereby stabilising the zinc-deficient mono-
with studies showing co-expression of wild-type SOD1— meric species and exacerbating its toxicity in the presence of
which would be expected to ameliorate peroxynitrite dam- NO [44].
age—worsens disease progression [35], have made this a less Initial studies on zinc supplementation as a treatment for
likely hypothesis. The ability for H2O2 to induce peroxidation mutant SOD1 mice demonstrated an increase in the rate of
damage on cellular components and its central position within death [45]. A follow-up study however attributed this outcome
the antioxidant role of SOD1 provided another popular candi- to excessive zinc treatment leading to competition for copper
date for the SOD1-dependent toxic species. But, the same absorption that ultimately inhibited ceruloplasmin resulting in
limitations apply as for the peroxynitrite hypothesis where fatal anaemia [46]. An amended treatment programme using a
SOD1G85R mutations induce an ALS phenotype indepen- more modest zinc supplement concentration conversely dem-
dent of aberrant copper chemistry, with further evidence dem- onstrated an improved survival rate in the same mouse model
onstrating an inability to detect elevated levels of H2O2 in of ALS [46]. Whilst the zinc-deficient SOD1 induction of
mutant SOD1 models [36]. peroxynitrite-dependent pathology is widely supported, con-
Zinc is crucial for the structural integrity of SOD1 and tradictory evidence exists for the relevance of tyrosine nitra-
possesses a curiously weaker affinity than copper [37], tion in most SOD1 mutant models as well as human ALS
prompting the suggestion that a zinc-deficient species of cases [36, 47]. Considering the requirement of NO for
SOD1 may contribute to ALS (Fig. 1). It has been reported peroxynitrite production, the abatement of nitric oxide syn-
that mutant forms of SOD1 have a decreased affinity for the thase (NOS) generators of NO would be expected to amelio-
structural zinc atom in contrast to the wild-type species [38], rate disease symptoms. Yet the pharmacological inhibition of
presumably affecting the protein folding pattern and indirectly neuronal NOS [48] and genetic ablation of inducible NOS
affecting the catalytic activities of SOD1. Changes to the na- [49] failed to have an effect on the survival of SOD1 mutant
tive zinc binding site lead to increased zinc loss from SOD1 mice, casting some doubt upon the peroxynitrite hypothesis.
which subsequently becomes more relaxed and potentially
more accessible to substrates able to catalyse nitration reac-
tions. Additionally, through a shared bridging histidine resi- Copper-deficient SOD1
due [39], the consequence of zinc deficiency may be to change
the redox state of the bound copper which could facilitate As an alternative to zinc-deficient SOD1 driving toxicity in
protein nitration [37]. Neurofilaments and metallothioneins specific models of familial ALS, a role for a copper-deficient
bind readily to metals [40, 41] and are present in large cellular species has been mooted (Fig. 1). SOD1 has one of the
concentrations similar to SOD1, so it has been proposed that greatest affinities for copper, alongside metallothioneins, with
these proteins may act as a sink for zinc, leading to greater a reported ∼7000 times greater affinity for copper than zinc in
zinc-deficient SOD1 levels and therefore increased the case of wild-type SOD1 that was increased further in A4V
peroxynitrite-dependent nitration reactions [37, 42]. Due to SOD1 mutants [43]. As previously mentioned, the SOD1 loss-
reported similarities in the binding affinity for zinc between of-function hypothesis was rejected based on an absence of
mutant SOD1 and the wild-type form, it has been contended ALS-like pathology in SOD1 knockout mice, demonstrating
that there may exist the capacity for wild-type SOD1 to be- that a loss of copper-dependent dismutase activity in isolation
come zinc deficient in sporadic cases of ALS as well [38, 43]. does not cause ALS [12]. Additionally, when endogenous
More recently, it was reported that zinc-deficient SOD1 is mouse SOD1 was knocked out in mice expressing the
required to induce motor neuron death with copper chelation metal-binding region SOD1G85R mutant—which exhibits
acting to protect motor neurons from nitric oxide-induced extremely poor metal-binding properties [50]—there was no

Fig. 1 Overview of various

forms of SOD1 relative to metal
state, the potential contribution of
metal-deficient SOD1 to ALS and
therapeutic opportunity to
attenuate toxicity of metal-
deficient SOD1
484 J Mol Med (2015) 93:481–487

difference in survival or age of symptom onset [51]. Work mutant SOD1 and a corresponding increase in fully metallated
performed on the wild-type-like SOD1G37R mutant de- holo form of the protein. Changes to the metal-deficient pool
scribed above conversely demonstrated dismutase activity correlated directly to changes in the copper-deficient pool. It
levels that were unchanged relative to wild-type SOD1, yet was proposed that the CuII(atsm) treatment delivered copper
induced an ALS-like phenotype when expressed in transgenic directly to the mutant SOD1, and this was supported by an
mice [52]. Subsequent work investigating the metallation sta- experiment in which the mice were treated with isotopically
tus of mutant SOD1 species found that whilst metal-binding labelled 65CuII(atsm). Due to conversion of the metal-deficient
region SOD1 mutants were severely metal deficient, wild- SOD1 to the highly stable holo form, total levels of mutant
type-like mutant SOD1 species exhibited normal activity SOD1 were increased in the CuII(atsm)-treated mice. Based
levels per equivalent copper but a detectable decrease in metal on the fact that the phenotype of these mice was improved by
binding capacities relative to wild-type control [7]. This was the CuII(atsm) treatment, despite an overall increase in levels
supported through computational analysis performed on the of mutant SOD1, it was concluded that the metal state of the
SOD1G37R mutant which showed that this point mutation, SOD1 is a greater determinant of the protein’s role in motor
which is not within the metal binding region, could also in- neuron death and the ALS-like phenotype of these animals
duce metal binding impairment leading to decreased copper than the mutant amino acid sequence per se. Interestingly,
affinity ahead of zinc based on free energy calculations [53]. the overall increase in SOD1 levels in the CuII(atsm)-treated
Several studies, utilising copper chelating compounds in mice also included an increase in levels of the protein detected
SOD1 mutant models of ALS intended to inhibit the purport- using antibodies selective for the misfolded species [61].
edly aberrant copper chemistry, have reported protection Taken together, these results indicate that improving the
against motor neuron loss with a corresponding improvement copper bioavailability within the spinal cord tissue of SOD1
in lifespan and locomotor function [54–56]. The mechanisms mutant mice and improving SOD1 metallation status are poten-
of action for therapeutic benefit suggested attenuation of cop- tial mechanisms for attenuating the ALS phenotype. Further-
per ion toxicity [55], decreased spinal cord copper ion levels more, since the unmetallated apo form of SOD1 is rapidly
and reduced lipid peroxidation [56], or decreased markers of turned over [64] and the fully metallated form is stable and
oxidative damage and inflammation [54]. Contrarily, previous likely to be non-toxic regardless of amino acid substitution
work on the successful imaging compound diacetylbis(4- mutations [65–68], it appears that a partially metal-deficient
methylthiosemicarbazonato)copperII [57] demonstrated that intermediate species may be responsible for the SOD1 toxic
under conditions of impaired mitochondrial transport chain gain of function. This could potentially explain the contradic-
function and elevated NADH levels, treatment with tory data in the literature reporting therapeutic benefits from
CuII(atsm) increased intracellular retention of copper [58]. both copper chelating compounds and Bcopper delivery
When used to treat a SOD1G93A mouse model of ALS, the agents^, both of which may be acting to shift the equilibrium
compound significantly delayed locomotor deficit onset and away from this potentially toxic copper-deficient species of
improved survival, decreased levels of peroxynitrite-induced SOD1 (Fig. 1). It is currently not clear why such a large pool
protein nitration and increased SOD1 activity in spinal cord of copper-deficient SOD1 would accumulate within the spinal
tissue [57]. Collectively, these studies indicated that a thera- cord tissue of SOD1G37R mice, but outcomes from the
peutic agent capable of increasing copper bioavailability could SOD1G93A mouse model have demonstrated that widespread
protect against mutant SOD1 toxicity, a possibility consistent and progressive impairment of intracellular copper trafficking
with previous in vitro studies which had described the poten- occurs due to expression of the mutant SOD1 and that these
tial for copper deficiency to promote SOD1 pathology via changes to copper trafficking can be detected at a presymptom-
increased misfolding and altered hydrophobicity [59, 60]. It atic stage [69]. Thus, it is apparent that copper homeostasis is
was however a more recent study [61] involving CuII(atsm) substantially altered by the expression of mutant SOD1 and that
treatment in a SOD1G37R mouse model of ALS that illustrat- therapeutically modulating copper bioavailability can attenuate
ed the potential link between copper-deficient SOD1 and ALS ALS-like symptoms in mutant SOD1 mice.
pathogenesis in vivo. Oral treatment with the compound again Definitive mechanistic data to support altered copper ho-
showed improved locomotor function and prolonged survival meostasis as a valid therapeutic basis for treating ALS in the
compared to untreated mice, concomitant with decreased mo- clinic, particularly sporadic cases of the disease, are yet to be
tor neuron death within the spinal cords of treated mice. Using reported. However, mutations to the copper transporter
a FTICR mass spectrometry technique [62, 63], it was shown ATP7A that lead to systemic copper deficiency are an
that zinc-deficient mutant SOD1 levels in both treated and established cause of the infantile-onset neurodegenerative
untreated mutant mice were relatively low (∼1 μM), in con- Menkes disease [70, 71], and some ATP7A missense muta-
trast to the substantially more abundant (∼60 μM) copper- tions have been shown to cause X-linked distal hereditary
deficient form of the protein. Treatment with CuII(atsm) in- motor neuropathies in the absence of systemic copper defi-
duced a significant decrease in the levels of metal-deficient ciency [72]. This suggests that motor neurons may be
J Mol Med (2015) 93:481–487 485

particularly sensitive to altered copper homeostasis, and this References

has been supported recently by a report describing a degener-
ative phenotype (involving progressive muscular atrophy, loss 1. Naganska E, Matyja E (2011) Amyotrophic lateral sclerosis—
of motor neuron cell bodies and denervation of the neuromus- looking for pathogenesis and effective therapy. Folia Neuropathol
cular junction) in mice caused by selective knockout of the 49(1):1–13
2. Kiernan MC, Vucic S, Cheah BC, Turner MR, Eisen A, Hardiman O,
Atp7a gene from motor neurons [73]. Although such an ex-
Burrell JR, Zoing MC (2011) Amyotrophic lateral sclerosis. Lancet
plicit role for altered copper homeostasis as a cause of ALS is 377(9769):942–955
not yet evident, these studies nonetheless indicate the funda- 3. Beleza-Meireles A, Al-Chalabi A (2009) Genetic studies of amyo-
mental requirement for copper in maintaining motor neuron trophic lateral sclerosis: controversies and perspectives. Amyotroph
Lateral Scler 10(1):1–14
functionality.
4. Rosen DR, Siddique T, Patterson D, Figlewicz DA, Sapp P, Hentati
A, Donaldson D, Goto J, O’Regan JP, Deng HX et al (1993)
Mutations in Cu/Zn superoxide dismutase gene are associated with
familial amyotrophic lateral sclerosis. Nature 362(6415):59–62
5. Perry JJ, Shin DS, Getzoff ED, Tainer JA (2010) The structural bio-
Concluding remarks
chemistry of the superoxide dismutases. Biochim Biophys Acta
1804(2):245–262
After more than 20 years of research since SOD1 mutations 6. Valentine JS, Doucette PA, Zittin Potter S (2005) Copper-zinc super-
were discovered as a definitive cause of ALS, the mechanisms oxide dismutase and amyotrophic lateral sclerosis. Annu Rev
Biochem 74:563–593
by which SOD1 cause ALS remain far from elucidated. Trans-
7. Hayward LJ, Rodriguez JA, Kim JW, Tiwari A, Goto JJ, Cabelli DE,
genic SOD1 animal models that develop robust ALS-like phe- Valentine JS, Brown RH Jr (2002) Decreased metallation and activity
notypes have been created and have been used extensively for in subsets of mutant superoxide dismutases associated with familial
preclinical drug development, but instead of providing clear amyotrophic lateral sclerosis. J Biol Chem 277(18):15923–15931
insight to the pathogenesis of ALS, these animal models have 8. Rodriguez JA, Valentine JS, Eggers DK, Roe JA, Tiwari A, Brown
RH Jr, Hayward LJ (2002) Familial amyotrophic lateral sclerosis-
often been the source of apparent confounding outcomes. But associated mutations decrease the thermal stability of distinctly
progress in being made and the role for the metal state of metallated species of human copper/zinc superoxide dismutase. J
SOD1 in ALS is one area of research providing new opportu- Biol Chem 277(18):15932–15937
nity for the development of effective therapeutic options. It is 9. Tiwari A, Hayward LJ (2003) Familial amyotrophic lateral sclerosis
mutants of copper/zinc superoxide dismutase are susceptible to disul-
becoming clear that the metal state of SOD1 can be modulated fide reduction. J Biol Chem 278(8):5984–5992
via therapeutic intervention and that the metal state of SOD1 10. Longo VD, Gralla EB, Valentine JS (1996) Superoxide dismutase
may be a greater determinant of the protein’s elusive toxic activity is essential for stationary phase survival in Saccharomyces
gain of function than the disease-causing mutations them- cerevisiae. Mitochondrial production of toxic oxygen species in vivo.
J Biol Chem 271(21):12275–12280
selves. Further to this, evidence for a toxic gain of function 11. Reaume AG, Elliott JL, Hoffman EK, Kowall NW, Ferrante RJ,
and structural instability in metal-deficient wild-type SOD1 Siwek DF, Wilcox HM, Flood DG, Beal MF, Brown RH Jr et al
indicates a mechanism by which SOD1 contributes to cases (1996) Motor neurons in Cu/Zn superoxide dismutase-deficient mice
of ALS that do not involve SOD1 mutations. Whether or not develop normally but exhibit enhanced cell death after axonal injury.
Nat Genet 13(1):43–47
metal-deficient wild-type SOD1 contributes to the pathogen- 12. Ho YS, Gargano M, Cao J, Bronson RT, Heimler I, Hutz RJ (1998)
esis of sporadic ALS is yet to be established, but if proven Reduced fertility in female mice lacking copper-zinc superoxide dis-
could provide opportunity to treat all cases of ALS, not just mutase. J Biol Chem 273(13):7765–7769
mutant SOD1 cases, using drugs designed to improve the 13. Tokuda E, Okawa E, Watanabe S, Ono S (2014) Overexpression of
metallothionein-I, a copper-regulating protein, attenuates intracellular
protein’s metal state. copper dyshomeostasis and extends lifespan in a mouse model of
amyotrophic lateral sclerosis caused by mutant superoxide dismut-
ase-1. Hum Mol Genet 23(5):1271–1285
Acknowledgments This work was supported by funds from the Na- 14. Swarup V, Julien JP (2011) ALS pathogenesis: recent insights from
tional Health and Medical Research Council, the Australian Research genetics and mouse models. Prog Neuropsychopharmacol Biol
Council, and the University of Melbourne. Psychiatry 35(2):363–369
15. Lalonde R, Dumont M, Paly E, London J, Strazielle C (2004)
Characterization of hemizygous SOD1/wild-type transgenic mice
Conflict of interest Collaborative Medicinal Development LLC has with the SHIRPA primary screen and tests of sensorimotor function
licenced intellectual property pertaining to CuII(atsm) from the University and anxiety. Brain Res Bull 64(3):251–258
of Melbourne. JBH, ARW and PJC all hold current appointments at the 16. Graffmo KS, Forsberg K, Bergh J, Birve A, Zetterstrom P, Andersen
University of Melbourne and ARW is one of the named inventors on the PM, Marklund SL, Brannstrom T (2013) Expression of wild-type
licenced material. human superoxide dismutase-1 in mice causes amyotrophic lateral
sclerosis. Hum Mol Genet 22(1):51–60
Open Access This article is distributed under the terms of the Creative 17. Watanabe M, Dykes-Hoberg M, Culotta VC, Price DL, Wong PC,
Commons Attribution License which permits any use, distribution, and Rothstein JD (2001) Histological evidence of protein aggregation in
reproduction in any medium, provided the original author(s) and the mutant SOD1 transgenic mice and in amyotrophic lateral sclerosis
source are credited. neural tissues. Neurobiol Dis 8(6):933–941
486 J Mol Med (2015) 93:481–487

18. Forsberg K, Andersen PM, Marklund SL, Brannstrom T (2011) Glial 35. Jaarsma D, Haasdijk ED, Grashorn JA, Hawkins R, van Duijn W,
nuclear aggregates of superoxide dismutase-1 are regularly present in Verspaget HW, London J, Holstege JC (2000) Human Cu/Zn super-
patients with amyotrophic lateral sclerosis. Acta Neuropathol 121(5): oxide dismutase (SOD1) overexpression in mice causes mitochondri-
623–634 al vacuolization, axonal degeneration, and premature motoneuron
19. Shibata N, Hirano A, Kobayashi M, Siddique T, Deng HX, Hung death and accelerates motoneuron disease in mice expressing a famil-
WY, Kato T, Asayama K (1996) Intense superoxide dismutase-1 ial amyotrophic lateral sclerosis mutant SOD1. Neurobiol Dis 7(6 Pt
immunoreactivity in intracytoplasmic hyaline inclusions of familial B):623–643
amyotrophic lateral sclerosis with posterior column involvement. J 36. Bruijn LI, Beal MF, Becher MW, Schulz JB, Wong PC, Price DL,
Neuropathol Exp Neurol 55(4):481–490 Cleveland DW (1997) Elevated free nitrotyrosine levels, but not
20. Chattopadhyay M, Durazo A, Sohn SH, Strong CD, Gralla EB, protein-bound nitrotyrosine or hydroxyl radicals, throughout amyo-
Whitelegge JP, Valentine JS (2008) Initiation and elongation in fibril- trophic lateral sclerosis (ALS)-like disease implicate tyrosine nitra-
lation of ALS-linked superoxide dismutase. Proc Natl Acad Sci U S tion as an aberrant in vivo property of one familial ALS-linked su-
A 105(48):18663–18668 peroxide dismutase 1 mutant. Proc Natl Acad Sci U S A 94(14):
21. Furukawa Y, Kaneko K, Yamanaka K, O’Halloran TV, Nukina N 7606–7611
(2008) Complete loss of post-translational modifications triggers fi- 37. Crow JP, Sampson JB, Zhuang Y, Thompson JA, Beckman JS (1997)
brillar aggregation of SOD1 in the familial form of amyotrophic Decreased zinc affinity of amyotrophic lateral sclerosis-associated
lateral sclerosis. J Biol Chem 283(35):24167–24176 superoxide dismutase mutants leads to enhanced catalysis of tyrosine
22. Basso M, Massignan T, Samengo G, Cheroni C, De Biasi S, Salmona nitration by peroxynitrite. J Neurochem 69(5):1936–1944
M, Bendotti C, Bonetto V (2006) Insoluble mutant SOD1 is partly 38. Lyons TJ, Liu H, Goto JJ, Nersissian A, Roe JA, Graden JA, Cafe C,
oligoubiquitinated in amyotrophic lateral sclerosis mice. J Biol Chem Ellerby LM, Bredesen DE, Gralla EB et al (1996) Mutations in
281(44):33325–33335 copper-zinc superoxide dismutase that cause amyotrophic lateral
23. Cheroni C, Peviani M, Cascio P, Debiasi S, Monti C, Bendotti C sclerosis alter the zinc binding site and the redox behavior of the
(2005) Accumulation of human SOD1 and ubiquitinated deposits in protein. Proc Natl Acad Sci U S A 93(22):12240–12244
the spinal cord of SOD1G93A mice during motor neuron disease 39. Strange RW, Hough MA, Antonyuk SV, Hasnain SS (2012)
progression correlates with a decrease of proteasome. Neurobiol Structural evidence for a copper-bound carbonate intermediate in
Dis 18(3):509–522 the peroxidase and dismutase activities of superoxide dismutase.
24. Takeuchi H, Kobayashi Y, Yoshihara T, Niwa J, Doyu M, Ohtsuka K, PLoS One 7(9):e44811
Sobue G (2002) Hsp70 and Hsp40 improve neurite outgrowth and 40. Pierson KB, Evenson MA (1988) 200 Kd neurofilament protein
suppress intracytoplasmic aggregate formation in cultured neuronal binds Al, Cu and Zn. Biochem Biophys Res Commun 152(2):598–
cells expressing mutant SOD1. Brain Res 949(1–2):11–22 604
25. Weisberg SJ, Lyakhovetsky R, Werdiger AC, Gitler AD, Soen Y, 41. Shaw CF 3rd, Savas MM, Petering DH (1991) Ligand substitution
Kaganovich D (2012) Compartmentalization of superoxide dismut- and sulfhydryl reactivity of metallothionein. Methods Enzymol 205:
ase 1 (SOD1G93A) aggregates determines their toxicity. Proc Natl 401–414
Acad Sci U S A 109(39):15811–15816 42. Crow JP, Ye YZ, Strong M, Kirk M, Barnes S, Beckman JS (1997)
26. Karch CM, Prudencio M, Winkler DD, Hart PJ, Borchelt DR (2009) Superoxide dismutase catalyzes nitration of tyrosines by
Role of mutant SOD1 disulfide oxidation and aggregation in the peroxynitrite in the rod and head domains of neurofilament-L. J
pathogenesis of familial ALS. Proc Natl Acad Sci U S A 106(19): Neurochem 69(5):1945–1953
7774–7779 43. Trumbull KA, Beckman JS (2009) A role for copper in the toxicity of
27. Zetterstrom P, Stewart HG, Bergemalm D, Jonsson PA, Graffmo KS, zinc-deficient superoxide dismutase to motor neurons in amyotrophic
Andersen PM, Brannstrom T, Oliveberg M, Marklund SL (2007) lateral sclerosis. Antioxid Redox Signal 11(7):1627–1639
Soluble misfolded subfractions of mutant superoxide dismutase-1s 44. Sahawneh MA, Ricart KC, Roberts BR, Bomben VC, Basso M, Ye
are enriched in spinal cords throughout life in murine ALS models. Y, Sahawneh J, Franco MC, Beckman JS, Estevez AG (2010) Cu,
Proc Natl Acad Sci U S A 104(35):14157–14162 Zn-superoxide dismutase increases toxicity of mutant and zinc-
28. Brotherton TE, Li Y, Glass JD (2012) Cellular toxicity of mutant deficient superoxide dismutase by enhancing protein stability. J
SOD1 protein is linked to an easily soluble, non-aggregated form Biol Chem 285(44):33885–33897
in vitro. Neurobiol Dis 49:49–56 45. Groeneveld GJ, de Leeuw van Weenen J, van Muiswinkel FL,
29. Vonk WI, Wijmenga C, Berger R, van de Sluis B, Klomp LW (2010) Veldman H, Veldink JH, Wokke JH, Bar PR, van den Berg LH
Cu, Zn superoxide dismutase maturation and activity are regulated by (2003) Zinc amplifies mSOD1-mediated toxicity in a transgenic
COMMD1. J Biol Chem 285(37):28991–29000 mouse model of amyotrophic lateral sclerosis. Neurosci Lett
30. Ding F, Dokholyan NV (2008) Dynamical roles of metal ions and the 352(3):175–178
disulfide bond in Cu, Zn superoxide dismutase folding and aggrega- 46. Ermilova IP, Ermilov VB, Levy M, Ho E, Pereira C, Beckman JS
tion. Proc Natl Acad Sci U S A 105(50):19696–19701 (2005) Protection by dietary zinc in ALS mutant G93A SOD trans-
31. Cleveland DW, Rothstein JD (2001) From Charcot to Lou Gehrig: genic mice. Neurosci Lett 379(1):42–46
deciphering selective motor neuron death in ALS. Nat Rev Neurosci 47. Beal MF, Ferrante RJ, Browne SE, Matthews RT, Kowall NW,
2(11):806–819 Brown RH Jr (1997) Increased 3-nitrotyrosine in both sporadic and
32. Beckman JS, Estevez AG, Crow JP, Barbeito L (2001) Superoxide familial amyotrophic lateral sclerosis. Ann Neurol 42(4):644–654
dismutase and the death of motoneurons in ALS. Trends Neurosci 48. Dawson TM (2000) New animal models for Parkinson’s disease. Cell
24(11 Suppl):S15–S20 101(2):115–118
33. Borchelt DR, Lee MK, Slunt HS, Guarnieri M, Xu ZS, Wong PC, 49. Son M, Fathallah-Shaykh HM, Elliott JL (2001) Survival in a trans-
Brown RH Jr, Price DL, Sisodia SS, Cleveland DW (1994) genic model of FALS is independent of iNOS expression. Ann
Superoxide dismutase 1 with mutations linked to familial amyotro- Neurol 50(2):273
phic lateral sclerosis possesses significant activity. Proc Natl Acad Sci 50. Tiwari A, Xu Z, Hayward LJ (2005) Aberrantly increased hydropho-
U S A 91(17):8292–8296 bicity shared by mutants of Cu, Zn-superoxide dismutase in familial
34. Wang L, Gutmann DH, Roos RP (2011) Astrocyte loss of mutant amyotrophic lateral sclerosis. J Biol Chem 280(33):29771–29779
SOD1 delays ALS disease onset and progression in G85R transgenic 51. Bruijn LI, Houseweart MK, Kato S, Anderson KL, Anderson SD,
mice. Hum Mol Genet 20(2):286–293 Ohama E, Reaume AG, Scott RW, Cleveland DW (1998)
J Mol Med (2015) 93:481–487 487

Aggregation and motor neuron toxicity of an ALS-linked SOD1 mu- motor neurons and improves the phenotype of a transgenic mouse
tant independent from wild-type SOD1. Science 281(5384):1851– model of amyotrophic lateral sclerosis. J Neurosci 34(23):8021–8031
1854 62. Rhoads TW, Lopez NI, Zollinger DR, Morre JT, Arbogast BL, Maier
52. Wong PC, Pardo CA, Borchelt DR, Lee MK, Copeland NG, Jenkins CS, DeNoyer L, Beckman JS (2011) Measuring copper and zinc
NA, Sisodia SS, Cleveland DW, Price DL (1995) An adverse prop- superoxide dismutase from spinal cord tissue using electrospray mass
erty of a familial ALS-linked SOD1 mutation causes motor neuron spectrometry. Anal Biochem 415(1):52–58
disease characterized by vacuolar degeneration of mitochondria. 63. Rhoads TW, Williams JR, Lopez NI, Morre JT, Bradford CS,
Neuron 14(6):1105–1116 Beckman JS (2013) Using theoretical protein isotopic distributions
53. Milardi D, Pappalardo M, Grasso DM, La Rosa C (2010) Unveiling to parse small-mass-difference post-translational modifications via
the unfolding pathway of FALS associated G37R SOD1 mutant: a mass spectrometry. J Am Soc Mass Spectrom 24(1):115–124
computational study. Mol Biosyst 6(6):1032–1039 64. Di Noto L, Whitson LJ, Cao X, Hart PJ, Levine RL (2005)
54. Petri S, Calingasan NY, Alsaied OA, Wille E, Kiaei M, Friedman JE, Proteasomal degradation of mutant superoxide dismutases linked to
Baranova O, Chavez JC, Beal MF (2007) The lipophilic metal che- amyotrophic lateral sclerosis. J Biol Chem 280(48):39907–39913
lators DP-109 and DP-460 are neuroprotective in a transgenic mouse 65. Kitamura F, Fujimaki N, Okita W, Hiramatsu H, Takeuchi H (2011)
model of amyotrophic lateral sclerosis. J Neurochem 102(3):991– Structural instability and Cu-dependent pro-oxidant activity acquired
1000 by the apo form of mutant SOD1 associated with amyotrophic lateral
55. Hottinger AF, Fine EG, Gurney ME, Zurn AD, Aebischer P (1997) sclerosis. Biochemistry 50(20):4242–4250
The copper chelator d-penicillamine delays onset of disease and ex- 66. Das A, Plotkin SS (2013) Mechanical probes of SOD1 predict sys-
tends survival in a transgenic mouse model of familial amyotrophic tematic trends in metal and dimer affinity of ALS-associated mutants.
lateral sclerosis. Eur J Neurosci 9(7):1548–1551 J Mol Biol 425(5):850–874
56. Tokuda E, Ono S, Ishige K, Watanabe S, Okawa E, Ito Y, Suzuki T 67. Banci L, Bertini I, D’Amelio N, Libralesso E, Turano P, Valentine JS
(2008) Ammonium tetrathiomolybdate delays onset, prolongs sur- (2007) Metalation of the amyotrophic lateral sclerosis mutant glycine
vival, and slows progression of disease in a mouse model for amyo- 37 to arginine superoxide dismutase (SOD1) apoprotein restores its
trophic lateral sclerosis. Exp Neurol 213(1):122–128 structural and dynamical properties in solution to those of metalated
57. Soon CP, Donnelly PS, Turner BJ, Hung LW, Crouch PJ, Sherratt wild-type SOD1. Biochemistry 46(35):9953–9962
NA, Tan JL, Lim NK, Lam L, Bica L et al (2011) Diacetylbis(N(4)- 68. Banci L, Bertini I, Boca M, Girotto S, Martinelli M, Valentine JS,
methylthiosemicarbazonato) copper(II) (CuII(atsm)) protects against Vieru M (2008) SOD1 and amyotrophic lateral sclerosis: mutations
peroxynitrite-induced nitrosative damage and prolongs survival in and oligomerization. PLoS One 3(2):e1677
amyotrophic lateral sclerosis mouse model. J Biol Chem 286(51): 69. Tokuda E, Okawa E, Ono S (2009) Dysregulation of intracellular
44035–44044 copper trafficking pathway in a mouse model of mutant copper/zinc
58. Donnelly PS, Liddell JR, Lim S, Paterson BM, Cater MA, Savva MS, superoxide dismutase-linked familial amyotrophic lateral sclerosis. J
Mot AI, James JL, Trounce IA, White AR et al (2012) An impaired Neurochem 111(1):181–191
mitochondrial electron transport chain increases retention of the hyp- 70. Kaler SG (2011) ATP7A-related copper transport diseases-emerging
oxia imaging agent diacetylbis(4-methylthiosemicarbazonato)copperII. concepts and future trends. Nat Rev Neurol 7(1):15–29
Proc Natl Acad Sci U S A 109(1):47–52 71. Dierick HA, Ambrosini L, Spencer J, Glover TW, Mercer JF (1995)
59. Tiwari A, Liba A, Sohn SH, Seetharaman SV, Bilsel O, Matthews Molecular structure of the Menkes disease gene (ATP7A). Genomics
CR, Hart PJ, Valentine JS, Hayward LJ (2009) Metal deficiency 28(3):462–469
increases aberrant hydrophobicity of mutant superoxide dismutases 72. Kennerson ML, Nicholson GA, Kaler SG, Kowalski B, Mercer JF,
that cause amyotrophic lateral sclerosis. J Biol Chem 284(40):27746– Tang J, Llanos RM, Chu S, Takata RI, Speck-Martins CE et al (2010)
27758 Missense mutations in the copper transporter gene ATP7A cause X-
60. Ip P, Mulligan VK, Chakrabartty A (2011) ALS-causing SOD1 mu- linked distal hereditary motor neuropathy. Am J Hum Genet 86(3):
tations promote production of copper-deficient misfolded species. J 343–352
Mol Biol 409(5):839–852 73. Hodgkinson VL, Dale JM, Garcia ML, Weisman GA, Lee J, Gitlin
61. Roberts BR, Lim NK, McAllum EJ, Donnelly PS, Hare DJ, Doble JD, Petris MJ (2015) X-linked spinal muscular atrophy in mice
PA, Turner BJ, Price KA, Lim SC, Paterson BM et al (2014) Oral caused by autonomous loss of ATP7A in the motor neuron. J
treatment with CuII(atsm) increases mutant SOD1 in vivo but protects Pathol. doi:10.1002/path.4511