Many proteins associated with neurodegenerative diseases have poorly defined or unknown functions. α-Synuclein is one such protein which is associated with a range of diseases including Parkinson's disease. Now accepted as a metal-binding protein, α-synuclein's function could possibly be defined in relation to the binding of cofactors. It has been suggested recently that α-synuclein is able to reduce iron using copper as its catalytic centre. The consequence of this is that possibly the function of α-synuclein can now be defined. The evidence for this and the consequences for Parkinson's disease are discussed in the present review.
α-Synuclein is a protein expressed in the cytosol of neurons. It belongs to a family of similar proteins [1,2]. One of these proteins, β-synuclein is also expressed in the brain. α-Synuclein is a 14-kDa protein that currently has no clearly defined function. It is mostly known for its possible involvement in a range of diseases . In these diseases, α-synuclein changes its structure to one that can readily aggregate. Aggregates of α-synuclein commonly form deposits known as Lewy bodies which are found in a variety of diseases .
In Alzheimer's disease, α-synuclein was first known as the non-amyloid component because it was present in amyloid plaques . Although there has been little continued interest in α-synuclein in Alzheimer's disease, the protein is more commonly associated with Parkinson's disease . Other diseases that are associated with α-synuclein include dementia with Lewy bodies and multiple system atrophy . Parkinson's disease differs slightly from the other diseases in that there has also been association of that disease with a range of other proteins including parkin and LRRK2 (leucine-rich repeat kinase 2) . These other proteins are mostly associated with either phosphorylation or protein degradation. However, in the case of α-synuclein, there is clear evidence from inherited forms of Parkinson's disease that this protein is a key player in the pathology. Changes in the promoter and five point mutations in the coding sequence for the protein have been identified in SNCA, the α-synuclein gene [9–11]. These mutations are clustered in the N-terminus of the protein, which clearly suggests that they interfere with a role played by the N-terminus of the protein in its normal cellular role.
Suggested functions for α-synuclein
Parkinson's disease is largely a disease associated with loss of dopaminergic innervation of the striatum through death of neurons in the substantia nigra . Loss of this pathway leads to the motor disturbances associated with the disease. As a result of this central component of the disease, it has been very tempting to try to link the function of α-synuclein to the biology of dopamine [13,14]. These studies have mostly focused on the possible role of α-synuclein on dopamine packaging into vesicles and subsequent vesicle release from neurons . In particular, there is recent evidence that the protein acts as a chaperone during SNARE (soluble N-ethylmaleimide-sensitive fusion protein-attachment protein receptor) complex formation [16,17].
The general approach to identify functions for α-synuclein (and proteins involved in other neurodegenerative diseases) has been to look for cellular processes that are disrupted either as a result of loss of expression of the protein or when the protein aggregates. Such studies are somewhat flawed in their approach, as it is widely accepted that most cellular processes are multifactorial, and changing any one protein is likely to disrupt or alter many such processes. Nevertheless, evidence has been proposed for a role of α-synuclein in such cellular processes as vesicle trafficking , neurotransmitter metabolism , neurogenesis , melanin synthesis , mitochondrial activities  and cell viability . There is also some evidence that α-synuclein expression is neuroprotective [24,25], and it is only under pathological conditions that its increased expression is detrimental to cells following its aggregation. Despite many different approaches and a range of suggested functions, no single function has emerged from the pack as more likely than any other. This is most likely because none of the proposed ‘functions’ actually assigns a specific molecular activity to the protein itself.
As with many proteins associated with neurodegeneration, much of the molecular details of the protein are hotly contested. Although there is not a lot of evidence for any single function, it was largely assumed that the normal structure of the protein was not a matter of contention. The protein has been thought to be monomeric and disorganized when free in the cytosol, but when associated with the membrane the protein gained small α-helical content . There has been a suggestion recently that the protein is normally a tetramer [27,28] and that the formation of this largely helical complex prevents the protein from aggregating. In this model, failure of the protein to form tetramers would then allow the free monomeric α-synuclein to be drawn into aggregation processes leading to the generation of toxic oligomers or fibrils. There is also some evidence that its homologue β-synuclein can also form tetramers . Although some cell biology and structural biology studies have shown that tetramers of α-synuclein can form, there has also been work contesting that the tetrameric form is not the main species of the protein in cells . Thus, at present, the relevance of this tetrameric synuclein is unclear. However, in searching for a possible function for this protein, it has to be considered that the functional form might be tetrameric or that both the monomeric and the tetrameric forms might have different functions.
The co-localization of α- and β-synuclein also suggests that there might be some relationship between the function of both proteins. In both the mouse brain and the human substantia nigra, α-synuclein mRNA decreases and β-synuclein mRNA increases with age . In contrast with control patients, there is a dramatic increase in α-synuclein and decrease in β-synuclein mRNA levels in the substantia nigra of Parkinson's disease, diffuse Lewy body disease and a Lewy body variant of Alzheimer's disease , as well as the cortex in dementia with Lewy bodies . Although there has been some disagreement with these findings, such changes in expression have been confirmed in the majority of studies [34,35]. This switch in relative synuclein transcript levels with disease suggests that the balance of α-synuclein and β-synuclein expression may be important, which is supported by several studies [36,37]. An increased level of expression of β-synuclein has been reported to inhibit the aggregation of α-synuclein . Point mutations of β-synuclein are also found in cases of dementia with Lewy bodies  and thus verify that altering β-synuclein can result in α-synuclein aggregation and disease. Despite increasing evidence linking the two proteins in terms of expression, activity and changes in pathology, the function of β-synuclein is even less well understood than that of α-synuclein. It is quite likely that there is a relationship between the two, and determining the function of α-synuclein might help to identify the function of β-synuclein.
A number of proteins associated with neurodegenerative diseases have been shown to bind metals. These include the amyloid precursor protein, β-amyloid and the prion protein . It is therefore not surprising that the potential metal binding of α-synuclein was investigated [41,42]. Metals often play a role as cofactors for proteins. They are necessary either for structural stability, such as the zinc cofactor of superoxide dismutase, or for a role in catalytic activities of enzymes. However, in the study of α-synuclein, metals were first investigated in terms of their potential to accelerate the aggregation of the protein . Copper was the first metal to be shown to bind to α-synuclein. Since then some studies have shown that a number of metals can bind to the protein. The co-ordination of copper has been investigated and two potential sites of metal interaction have been defined. In particular, a site at the N-terminus has been shown to be the high-affinity site . There is some disagreement with regard to the stoichiometry and affinity of copper binding. Our own previous study has identified all three synucleins (α-, β- and γ-synuclein) as copper-binding proteins, and suggested a single copper-binding site in the N-terminus with high affinity that has two potential modes of co-ordination .
Chief among other metals that have been identified as binding to α-synuclein is iron . Iron is a more interesting metal with regard to synucleinopathies because it has been reported for some years that iron levels are altered in Parkinson's disease. Iron binding to synuclein was originally suggested to be relatively low affinity. Our studies suggest that this is not the case and that α-synuclein can bind to two atoms of iron with low micromolar affinity . Additionally, α-synuclein can bind both copper and iron simultaneously, as shown by isothermal titration calorimetric studies of iron binding to copper-saturated protein .
The reason for the investigation into metal binding to a protein is the possible insight that it provides for its function. As metals are often cofactors associated with catalytic activity, this could provide a key piece of evidence linking the protein to a biochemical function. Copper is normally a cofactor associated with electron transfer in oxidation and reduction reactions . Key to this is the ability of copper, bound to the protein in question, to cycle reversibly between oxidized and reduced forms. We have studied α-synuclein using cyclic voltammetry, an electrochemical technique that looks at the ability of a protein–metal complex to cycle between oxidized and reduced forms of the bound metal. α-Synuclein with copper bound showed a strong ability to cycle between oxidized and reduced forms. This implies that, in terms of copper binding to the protein, it could play a role in catalytic processes of electron donation.
The reduction of cellular iron from Fe(III) to Fe(II) is an important process as it provides the form of iron commonly used by other enzymes such as tyrosine hydroxylase, a key enzyme in the pathway of dopamine synthesis. We used a standard assay system to study iron reduction and found that both purified recombinant protein and cells overexpressing α-synuclein were able to reduce Fe(III). Using the purified recombinant protein we were able to carry out a kinetic study of this enzymatic activity and determined the Km and Vmax. We also looked at the ratio of Fe(III) to Fe(II) in cells overexpressing α-synuclein. There was a clear increase in the amount of Fe(II) in the cells. This supports the notion that increased α-synuclein levels result in increased products of a ferrireductase reaction. Taken together, data from these studies supports the notion that α-synuclein has the function of a cellular iron-reducing protein or a ferrireductase  (Figure 1).
α-Synuclein as a ferrireductase
In vivo relevance
The consequence of this finding, for both normal health and the conditions leading to neurodegeneration, remain unclear. It is very unlikely that α-synuclein is the only ferrireductase in cells, and therefore the question is to what extent does α-synuclein contribute to this cellular process. In the diseased state there are two possibilities regarding how alteration of ferrireductase activity could contribute to disease. The first is that increased expression of α-synuclein, as is common in Parkinson's disease, results in high levels of Fe(II) which then initiates oxidative damage through Fenton chemistry. Alternatively, aggregation of α-synuclein might result in loss of the conversion of Fe(III) to Fe(II), which might affect neuronal viability through indirect means (Figure 2). There is evidence of altered iron levels and altered ratios of Fe(III) to Fe(II) in Parkinson's disease [47,48]. The interaction of iron and neuromelanin has been suggested as a possible trigger for neuronal cell death . Assessing the level of ferrireductase activity in post-mortem brain tissue will prove a significant challenge. The activity we have measured is quite labile and is lost from cell extracts stored at low temperatures within a couple of weeks. Therefore assessing difference in ferrireductase activity in post-mortem tissue may not be viable. However, assuming some relationship can be determined for the level of α-synuclein expressed and ferrireductase activity in post-mortem brain areas such as the substantia nigra, then it should be possible to determine whether this relationship is altered in Parkinson's disease or other synucleinopathies. If this is not possible, then alternative means to determine in vivo relevance will need to be investigated, such as transgenic rodent models.
Ferrireductase activity and disease
The use of standard biochemical techniques has provided evidence that α-synuclein acts as a ferrireductase. This provides the first clear molecular activity for α-synuclein. Although other functions for the protein have been suggested, the evidence for these is largely inferred from the alteration of other cellular mechanisms, and as such is only indirect evidence. The identification of the function of a protein associated with a range of neurodegenerative diseases opens up potential investigations of whether altering that activity has an impact on either disease onset or progression. Given that it is already known that iron metabolism is altered in Parkinson's disease, there is a strong possibility that altered iron metabolism is due to changes in α-synuclein. Whatever future research uncovers, the identification of α-synuclein's ferrireductase activity is an advance of significant importance.
5th Conference on Advances in Molecular Mechanisms Underlying Neurological Disorders: A joint Biochemical Society/European Society for Neurochemistry Focused Meeting held at the University of Bath, U.K., 23–26 June 2013. Organized and Edited by Marcus Rattray (University of Bradford, U.K.) and Rob Williams (University of Bath, U.K.).
I thank Hazel Roberts for proofreading the review before submission.