U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Madame Curie Bioscience Database [Internet]. Austin (TX): Landes Bioscience; 2000-2013.

Cover of Madame Curie Bioscience Database

Madame Curie Bioscience Database [Internet].

Show details

α-Synuclein Physiology and Membrane Binding

and .

An amphipathic α-helical domain in the N-terminus of α-synuclein (AS) mediates binding to phospholipid membranes. This domain is structurally similar to the class A2 lipid-binding motif described for the exchangeable apolipoproteins. AS is unstructured in aqueous solution, but shifts to a unique α11/3-helical conformation in the presence of phospholipid vesicles or SDS. In addition, AS binds fatty acids, particularly those with long polyunsaturated acyl chains (PUFAs), and these interactions can induce irreversible multimerization of the protein. Biochemically, AS functions as an inhibitor of phospholipases D1 and D2 (PLD1 and PLD2), which catalyze the cleavage of phosphatidylcholine (PC) to generate phosphatidic acid (PA) and free choline. AS also regulates the activity of the plasma membrane dopamine transporter, probably via a mechanism involving internalization and sequestration of the transporter protein. Still other evidence links AS to the maintenance of vesicular pools in the presynapse. This review considers how these distinct functions may relate to the unique membrane interactions of AS, and the significance of such interactions for AS structure and function.

Introduction

Maroteaux and Scheller first isolated synuclein as a protein associated with cholinergic vesicles in the electroplaques of the electric fish, Torpedo californica.1 Cloning of related sequences in rat brain led to the identification of two synuclein family members, α- and β-synuclein.2 Subsequently, a third member of the synuclein family (γ-synuclein) was identified as a marker for breast cancer.3 While α- and β-synucleins are primarily localized to the brain, γ-synuclein is localized more broadly, in tissues including brain, peripheral nerve, retina, and several tumors (reviewed in refs. 4, 5). The α-, β-, and γ- proteins are now known to be the products of 3 distinct genes in humans. The original Torpedo synuclein is most similar to human γ-synuclein.5

α-synuclein (AS) has been of particular interest recently because of its association with Parkinson's disease and other neurodegenerative disorders. The AS protein is very abundant in the vertebrate brain, and it localizes specifically to presynaptic terminals,2,6 suggesting an interaction with presynaptic vesicles. However, despite evidence (reviewed below) that AS is a membrane-binding protein, it does not stringently copurify with synaptic vesicles upon fractionation of brain tissue,6,7 and does not localize to the synaptic vesicle surface by electron microscopy.8

α-Synuclein Binds to Lipid Membranes

Structural Predictions Based on Synuclein Sequence

At the level of primary sequence, the synucleins are not closely related to any other known protein family. Their normal function has been quite elusive, and researchers have looked to secondary structure to identify functional domains. All synucleins bear a series of degenerate 11-residue repeats spanning the first 89 residues of the protein (fig. 1). These repeats are predictive of an amphipathic α-helical secondary structure.6,9 When this N-terminal domain is rendered as an α-helix, a striking similarity to the lipid-binding domains of exchangeable apolipoproteins is revealed, particularly to a structure identified as the class A2 amphipathic helix.9 When viewed end-on, the helix has a characteristic concentration of hydrophobic residues on one face, hydrophilic residues on the opposite face, and basic residues (Lys) at each hydrophobic/hydrophilic interface.5 The class A2 helix mediates high-affinity, reversible lipid interactions by the apolipoproteins,10 leading to the hypothesis that AS (like apolipoprotein) functions as a lipid-binding protein. In contrast to the highly conserved N-terminus, the C-terminus is less conserved, and is most notable for an abundance of acidic residues (Glu and Asp). Computer algorithms consistently identify this as an unstructured region of the molecule.

Figure 1. Domain structure of α-synuclein.

Figure 1

Domain structure of α-synuclein. The primary sequence of human α-synuclein is depicted roughly to scale. The 11-residue repeats are numbered from 1 to 7. The acidic C-terminus is also depicted, along with the relative positions of three (more...)

Specificity of Membrane Binding

Davidson et al were first to show that AS binds selectively to certain phospholipids organized as small unilamellar vesicles, or SUV.9 Mixtures of protein and vesicles were separated by size filtration chromatography. The protein coeluted in large complexes with vesicles containing acidic phospholipid, e.g., phosphatidic acid (PA), phosphatidylserine (PS), or phosphatidylinositol (PI), but not with vesicles composed only of phospholipids with a net neutral charge (PC or phosphatidylethanolamine, PE).9 The average secondary structure of AS in equilibrium with these SUVs was also measured by circular dichroism spectroscopy (fig. 2), and the results parallel the binding data: acidic phospholipids induce a conformational shift in the synuclein protein from random coil to >70% α-helix, while no change is observed with PC alone. This conformational shift was slightly abrogated by the PD-associated mutant A30P. Because this substitution introduces a proline residue into a helical domain, it would be predicted to create a small break in the helix, and slightly decrease overall helicity. The mutant A53T showed no such effect, and may have even increased helicity to a slight degree.

Figure 2. Lipid binding induces a conformational shift in α-synuclein which is abrogated by the PD-associated mutation A30P, but not A53T.

Figure 2

Lipid binding induces a conformational shift in α-synuclein which is abrogated by the PD-associated mutation A30P, but not A53T. Mean residues ellipticity, [θ], for recombinant proteins (100μg/ml) is shown in 50mM phosphate buffer, (more...)

The observed requirement for PA or PS could indicate a significant electrostatic interaction between AS and phospholipid. However, the headgroups of PA and PS are known to have a relatively smaller cross-sectional area than that of PC, which could create packing defects in the membrane surface that allow stabilizing interactions between AS and the hydrophobic membrane interior. This might help to explain the preference of AS for small (˜25nm) versus large (˜125nm) unilameller vesicles,9 as the greater curvature of the small vesicle results in a lower density of phospholipid headgroups on the surface of the outer membrane leaflet. Consistent with this, Jo et al reported significant interaction of AS with large multilamellar vesicles, but only when the MLV were formed in the presence of AS,11 a condition that does not require AS to penetrate a preorganized membrane surface. More recently, Nuscher et al reported high exothermic heat values when AS was titrated with SUV, but not MLV.12 The heat release was greater than could be accounted for by protein folding alone. They concluded that AS has a high affinity for membrane bilayer packing defects, and hypothesized that AS binding increases lipid order, thereby stabilizing small vesicles. It has been further observed that AS can bind even to neutral phospholipid (PC) monolayers at reduced surface pressures (when phospholipids are less tightly packed), but not at the surface pressures typical of SUV or MLV (A. Seelig and J. George, unpublished data).

Other reports indicate that binding of AS to acidic vesicles is enhanced by the presence of brain PE. AS was also observed to bind PE fractions that were separated by thin-layer chromatography (TLC).11 These observations could indicate a specific affinity of AS for the ethanolamine headgroup of PE. An alternative interpretation is that the apparent affinity for PE results from variations in acyl chain content of the natural source PE used in these experiments, as compared to synthetic phospholipid controls. Unsaturated acyl groups can alter lipid packing, which presumably would affect AS binding to a target membrane. Affinity for TLC-purified PE might likewise be attributable to interactions with unsaturated acyl groups, as work from our laboratory demonstrates that AS is biochemically reactive with certain polyunsaturated lipids.13

Recently, Narayanan et al14 presented evidence that seemed to contradict the reported specificity of AS for acidic phospholipid. They used real-time equilibrium fluorescence measurements to demonstrate strong binding of AS to planar membranes, independent of phospholipid charge. However, this binding event was not associated with a helical shift in conformation, and seemed to be mediated by insertion of a hydrophobic protein domain into the bilayer rather than the extended amphipathic domain, thus representing a fundamentally different interaction than described in other reports. It is not clear why size filtration experiments fail to detect significant binding of AS to neutral membranes. It is possible that the interaction is reversed by a shift in the binding equilibrium as lipid-bound and free protein are separated on the sizing column, rendering this method less sensitive.

Structure of Membrane-Bound AS

Mutagenesis of the AS protein shows that each of the first 3 coding exons (residues 1-41, 43-55, and 56-102) is capable of mediating binding to vesicles, while residues 103-140 do not drive lipid binding and do not undergo any conformational change in the presence of phospholipids.15 Eliezer et al16 confirmed this structural mapping using high-resolution NMR to assign backbone resonances within the AS sequence, either in solution, or in the presence of SDS micelles. They found that in solution, AS is largely unstructured, with some propensity for transient helical structure, especially between residues 6-37. In the presence of SDS micelles, the structure of the C-terminus (residues 103-140) does not appear to change relative to the structure in solution. However, the N-terminus (residues 1-102) interacts with the detergent micelles and adopts a highly helical conformation. They repeated the experiments with AS bound to lipid vesicles. These preparations did not yield sufficient resolution for assignment of secondary structure in the N-terminus, probably due to the much larger size of the lipid vesicles relative to SDS micelles. However, they were able to confirm that the C-terminus does not interact with vesicles.

In further studies,17 this same group showed that the N-terminus of AS adopts a continuously helical conformation with a single break at residues 43-44, corresponding to the boundary between the first and second coding exons. They speculated that such a “hinge” could be necessary to allow AS binding to lipid surfaces of different curvature. They also observed that a continuous helical structure would require adoption of a slightly unwound α11/3 conformation (3.67 residues per turn as compared to 3.60 residues in a perfect α-helix) to maintain the amphipathicity of the helix. This deviation from an ideal α-helix creates strain in the backbone of the protein, which is energetically less favorable and might thus contribute to a capacity of AS to interact reversibly with membranes. Likewise, a series of threonine residues along the hydrophobic face of the helix might serve to weaken the interaction between the hydrophobic face of the helix and the oily interior of the membrane, thereby making the interaction more readily reversible.17

Jao et al recently provided direct evidence of the α11/3 helical model.18 They performed site-specific spin labeling of 47 distinct residues within the α-synuclein sequence, then employed EPR spectroscopy to measure the mobility at each site. In the absence of vesicles, spectra were consistent with loop or unfolded regions throughout. In the presence of 30% PS/70% PC vesicles, however, the spectra indicated that the N-terminal repeat domain forms a continuous helix with minimal tertiary contacts. Residues at equivalent positions within each repeat were found to have similar membrane proximity, indicating that the extended amphipathic helix lies parallel to the membrane surface. This could only occur if α-synuclein adopted a slightly unwound α11/3 helix predicted previously17 (see fig. 3). These authors also utilized spin-labeled phospholipids to estimate the depth of penetration of the helix into the membrane bilayer, and report that the core of the helix is located at a depth of ˜1-4Å, within the range estimated for exchangeable apolipoproteins19 with similar lipid-binding helices.

Figure 3. AS lipid-binding domain modeled as an α11/3 helix.

Figure 3

AS lipid-binding domain modeled as an α11/3 helix. Repeats 1-4 and 5-7 are projected as a helix with 11 residues per 3 turns (or 3.67 residues per turn). In this configuration, residues at similar positions within the 11mer repeat align at the (more...)

Lipid Interactions and α-Synuclein Function

Inhibition of Phospholipase D

The first biochemical function ascribed to α-synuclein was inhibition of the enzyme phospholipase D. In general, PLD catalyzes the cleavage of PC to generate PA and free choline. PA influences curvature of intracellular membranes20 and also serves to recruit certain specific proteins to the membrane, and thus has the potential to regulate a variety of membrane events (endocytosis, vesicle fusion, membrane budding, reviewed in 21). The PLD1 and PLD2 isozymes differ dramatically in their regulation and intracellular localization. PLD1 has low basal activity that is stimulated by PKC and by a variety of monomeric G proteins (ADP-ribosylation factor, Rho, Rac, and Cdc42)22. It is localized primarily to the perinuclear region (endoplasmic reticulum, Golgi apparatus, and late endosomes), where it is thought to promote formation of coated vesicles and regulate intracellular vesicular traffic.23 In contrast, purified PLD2 has high basal activity; its regulation in tissue seems to be mediated primarily by inhibition. When influenza-epitope-tagged PLD2 constructs were microinjected into quiescent rat embryo fibroblasts, PLD2 localized to the plasma membrane, where it induced formation of some irregular projections at the cell surface. Upon serum stimulation, filopodia were observed continuously across the cell surface, and a portion of the PLD2 protein was relocalized to a submembranous compartment,23 suggesting a role for PLD2 in endocytosis and in rearrangements of the actin cytoskeleton.

In 1998, AS was reported to regulate the activity of the phospholipase D isozyme PLD2.24 Jenco et al had previously purified PLD2 protein from bovine brain, and noted that the pure protein had high constitutive activity, while its activity in tissue was barely detectable.24 This observation provoked a search for endogenous inhibitors of the enzyme, leading to the biochemical purification of α and β-synuclein proteins from bovine brain. These two proteins were assayed for their ability to inhibit PLD2 and PLD1. Because PLD1 has low basal activity, but can be stimulated by monomeric G proteins (Rho and ARF family members) and by PKC in vitro, PLD1 activity was assayed in the presence of ARF. Under these conditions, α- and β-synucleins were potent inhibitors of PLD2 (Ki ˜10nM) but not of PLD1 (Ki >50μM).24

In a different set of experiments, Ahn et al observed inhibition of both PLD1 and PLD2 in human embryonic kidney cells (HEK 293).25 They stimulated PLD1 activity with the tyrosine-phosphatase inhibitor pervanadate, and observed lower activity in cells expressing AS, A53T, or A30P. Exogenous PLD2 activity was also inhibited by AS and PD-mutants in these experiments. These studies suggest that AS can inhibit either PLD1 or PLD2, but that inhibition of PLD1 may depend upon the means by which its activity is stimulated. This issue is less relevant for PLD2, which is constitutively active in vitro.25

The stimulatory effects of the tyrosine phosphatase inhibitor pervanadate suggested that PLD inhibition might be modulated by tyrosine phosphorylation. Ahn et al mutated each of the four tyrosine residues in α-synuclein (at positions 39, 125, 133, and 136) to phenylalanine, a change which precludes phosphorylation and mimics dephosphorylation.25 Only Y125F altered α-synuclein activity, increasing the inhibition of PLD, suggesting that tyrosine-phosphorylation at this site might relieve inhibition of PLD. A similar role has been demonstrated for serine phosphorylation. Pronin et al found that α- and β-synuclein proteins are targets of protein-coupled receptor kinases (GRKS), and mapped phosphorylation to S129 of α-synuclein and S118 of β-synuclein (homologous sites within the two proteins).26 GRK-phosphorylated α- and β-synuclein were significantly less potent inhibitors of PLD2 activity in vitro. Extending this result, Payton et al generated phosphorylation mimic mutants by introducing negatively charged residues in place of candidate phosphorylation sites.27 The mutations (S129E and Y125D/Y136D) essentially abolished PLD2 inhibition by α-synuclein, as did phosphorylation with the tyrosine kinase FYN and the serine/threonine kinase casein kinase II (which also targets serine 129).

Ahn et al also showed coimmunoprecipitation of PLD1 and PLD2 by AS, and mapped the protein domains responsible for the interaction between AS and PLD1. AS binds to a fragment of PLD1 (residues 1-331) containing phox and pleckstrin homology domains, while PLD1 binds both to the AS N-terminus (residues 1-60) and to the hydrophobic NAC domain (residues 61-95).25 In complementary work, our laboratory mapped the domains of AS required for PLD2 inhibition.27 We found that deletions of either exon 4 (residues 56-102) or exon 6 (residues 130-140) of AS could abolish its inhibitory activity. Taken together, and assuming that the mechanisms of inhibition are similar for PLD1 and PLD2, it appears that the central region of AS (containing the amphipathic repeats and the NAC domains) may directly interact with PLD, while the C-terminus plays a modulatory role that is influenced by phosphorylation.

Interestingly, other proteins found to be potent inhibitors of PLD activity include synaptic proteins like AP3/AP180,28 synaptojanin,29 and amphiphysin I and II.30 It is hypothesized that their activity is required for regulation of endocytosis, specifically the uncoating of clathrin-coated vesicles prior to their fusion with target membrane, an activity necessary for the rapid recycling of synaptic vesicles at the presynaptic nerve terminal.30

Regulation of Vesicular Pools in the Presynaptic Terminal

Consistent with a biochemical function in the regulation of vesicle traffic, both physiological and anatomical evidence implicate AS in the regulation of synaptic vesicle populations within the presynaptic terminal. In 2000, Abeliovich et al published the first description of α-Syn -/- mice.31 These animals were fertile and viable. Despite the known association of human AS mutations with nigrostriatal degeneration, 3-6 week-old α-Syn -/- mice did not differ from wild-type animals in numbers of dopaminergic neurons in the substantia nigra, density of nigral projections to the striatum, morphology and biochemical properties of striatal neurons and glia, or general striatal architecture, nor were gross morphological deficits observed in other brain regions. Striatal slice preparations from α-Syn -/- mice displayed normal DA release and reuptake after 1 or 10 pulses of a 20Hz stimulus. However, the α-Syn -/- mice exhibited more rapid recovery of DA release in a paired-stimulus paradigm. Striatal dopamine levels were measured in knockout mice, and levels of this neurotransmitter were reduced by 18% in striatum. The authors conclude that AS negatively regulates DA neurotransmission, perhaps via PLD2-dependent modulation of the size or recycling of the readily releasable (docked) pool of presynaptic vesicles.

Subsequently, Murphy et al examined the presynaptic morphology of cultured hippocampal neurons in which AS expression was partially reduced by treatment with antisense oligonucleotides.32 While the numbers of docked vesicles were unchanged by antisense treatment, there was a significant decrease in the distal (undocked) pool of synaptic vesicles in the AS-depleted cells versus controls (fig. 4). This was followed by concurring analysis of yet another α-Syn -/- mouse line by Cabin et al, who reported that AS knockout mice displayed 50% fewer undocked vesicles in hippocampal synapses than w.t. animals, while the numbers of docked vesicles remained unchanged.33 These mice had normal electrophysiological responses to stimuli expected to deplete only docked vesicles, but impaired responses to prolonged repetitive stimuli, which deplete both docked and undocked vesicles. These authors concluded that AS may be required for biogenesis or replenishment of undocked synaptic vesicles. Taken together, the available evidence strongly indicates that AS has a fundamental role in the maintenance and availability of synaptic vesicle pools in the presynaptic terminal.

Figure 4. α-synuclein antisense oligonucleotide treatment decreases the synaptic vesicle pool at the presynapse.

Figure 4

α-synuclein antisense oligonucleotide treatment decreases the synaptic vesicle pool at the presynapse. Left panels (a, c, e) are representative images of synapses from control hippocampal neurons. Right panels (b, d, f ) are images of representative (more...)

Potential Role As a Lipid Carrier or Chaperone

Other evidence indicates that AS may function as an intracellular lipid carrier or chaperone. Cole et al.34 found that HeLa cells treated with oleic acid accumulate abundant cytosolic lipid droplets, spherical particles with a core of neutral triglyceride and a surface monolayer of phospholipid. When these cells are transfected with different AS isoforms, wild-type and A53T proteins associate with the surface of these particles, while the A30P mutant does not. After washing away the oleic acid, the cells expressing wild-type AS maintain a population of stable lipid droplets with a surface coating of AS, while lipid droplets virtually disappear from the A53T and A30P-expressing cells (fig. 5). AS and β-synuclein were also shown to associate with the surface of lipid droplets in lipid-loaded hippocampal neurons.34 Given the structural relationship of AS to the exchangeable lipoproteins, the association of AS with cytosolic lipid particles suggests a unique pathway for intracellular shuttling of membrane lipids in neurons.

Figure 5. Localization of α-synuclein in cultured cells.

Figure 5

Localization of α-synuclein in cultured cells. HeLa cells were transfected with wild type (a, d, and g), A30P (b, e, and h), or A53T (c, f, and i) α-synuclein in control medium. After 36 h, control cells (a-c) were fixed and prepared for (more...)

Sharon et al have proposed that AS may function as a fatty acid binding protein.35 They showed that AS expression in a mesencephalic cell line results in altered levels of polyunsaturated fatty acids in both cytosolic and membrane fractions.36 In AS transfected cells, cytosolic levels of certain PUFA (20:5, 22:4, 22:5, 22:6) were significantly elevated, while levels of shorter or less unsaturated PUFA (18:2, 20:3, 20:4) were significantly increased in membrane fractions. These changes in PUFA content were correlated with increased membrane fluidity in the AS-transfected cells. Similar changes in fatty acid content were observed in α-Syn -/- mice as compared to the C57Bl parental strain, with a corresponding decrease in membrane fluidity. Such an activity could have profound effects on membrane fluidity and curvature, resulting in alterations in vesicle budding, fusion, or targeting. Alternatively, changes in membrane composition could represent a homeostatic mechanism to compensate for changes in membrane fluidity resulting from AS binding. Either way, a role for AS in lipid trafficking, storage, or membrane composition could have profound effects on membrane fluidity and curvature, resulting in alterations in vesicle budding, fusion, or targeting. In addition, Perrin et al have proposed that AS may also serve a protective function in biological membranes. AS reacts with fatty acid breakdown products in the membrane, causing irreversible multimerization of α-synuclein,13 but potentially quenching the feed-forward production of further damaging reactive species (R. Perrin, unpublished data). AS may thus function not only as a lipid carrier, but also as a chaperone that can buffer the propagation of reactive oxygen species generated by lipid peroxidation.

Dopamine Transporter Regulation

Reuptake of dopamine from the synaptic cleft is mediated by a dopamine transporter (DAT) on the presynaptic plasma membrane. This transporter can also mediate cellular uptake of the neurotoxin 1-methyl-4-phenyl pyridinium ion (MPP+),37,38 which has been associated with the development of parkinsonism in exposed individuals. In 2000, Lee et al reported direct binding of AS to the human dopamine transporter (hDAT), as determined by yeast two hybrid, coimmunoprecipitation, and GST fusion protein pull-down assays.39 Residues 58-108 of AS were sufficient to mediate binding to hDAT. This domain is unique to AS and is required for α-synuclein fibrillization; β-synuclein bears a deletion of this domain, and does not readily form fibrils.40 The complementary binding domain was localized to the C-terminal 15 amino acids (residues 598-620) of hDAT. Lee et al found that expression of AS could cause relocalization of hDAT to the plasma membrane in cotransfected HEK293 cells (a kidney cell line), and that formation of a complex between AS and hDAT caused an increase in DA transport across the plasma membrane. Vmax of the dopamine transporter increased significantly in Ltk- , COS-7, and HEK293 cells coexpressing α-synuclein, while Km was unaffected.39 They concluded that AS binds hDAT and increases its targeting to the plasma membrane, thereby increasing the rate of dopamine uptake.

Partially conflicting reports have been published by Wersinger et al, who found similar direct binding of AS to hDAT, but observed decreased (rather than increased) dopamine transport in cells cotransfected with AS and hDAT.41 Again, they saw no difference in binding of DA to the transporter, and Km did not change. However, they observed a significant decrease in Vmax. Since these papers used similar methodology, and observed similar values for Vmax in cells transfected with hDAT alone, it is difficult to reconcile the different effects. It has been proposed that subtle differences in experimental conditions (levels of AS expression, glucose concentration of culture media) may have contributed to the the contrasting results,41 but to date the issue has not been resolved.

In further studies, differential effects were found for the PD-associated mutations A30P and A53T. A30P and wild-type AS complex strongly with hDAT, but A30P inhibits its transporter activity to a greater degree. In contrast, A53T does not form complexes with hDAT, and does not modulate its activity.42 It was further shown that inhibition of hDAT by AS was dependent upon cell adhesion and sensitive to oxidative stress.43,44 Treatment of cultured cells with the neurotoxicant MPP+,43 or mild trypsinization43,44 sufficient to disrupt cell adhesion, led to increased hDAT localization to the cell surface, with a concomitant increase in plasma membrane dopamine transport. These data support a model whereby α-synuclein sequesters hDAT within the cell (a process dependent on an intact cytoskeleton) under normal conditions, thereby reducing dopamine reuptake. This negative regulation can be reversed by exposure to neurotoxicants or oxidative stress (factors associated with development of PD), increasing dopamine uptake and the potential for DA auto-oxidation within dopaminergic neurons.45

Functional Model

At this point, there is good evidence that α-synuclein can regulate two distinct biochemical processes: phospholipase D activity and dopamine transporter localization. There is not yet a direct link established between these two phenomena, beyond the observations that both PLD2 and hDAT have the capacity to cycle between the cell surface and a submembranous compartment. PLD2 is thought to mediate its own internalization, and might also contribute to internalization of hDAT, but this has yet to be tested. It seems likely that further functions will be ascribed to α-synuclein as our understanding of its biochemistry unfolds.

With that in mind, we propose the following general model of α-synuclein function. First, the presence of a highly conserved 11-residue repeat suggests that the true “native” (functional) state of AS is its membrane-bound, helical conformation. The degree of sequence conservation (e.g., 95% sequence identity between human and canary AS in the N-terminal 100 residues) is best explained in the context of strong selective pressures to maintain a lipid-dependent amphipathic helix; it is counterintuitive that the unstructured conformation typical of AS in solution should be the basis of such strict conservation. Helical AS interacts directly with a subset of intracellular membranes, physically altering the bilayer structure to influence trafficking of membrane between the cell interior and plasma membrane surface. The cytosolic face of this helix (also well-conserved) might interact with other proteins in a lipid-dependent manner, e.g., binding PLD2 or hDAT, or recruiting other specific factors required for AS activity. In each of these cases, the membrane environment controls the activity of AS, because the amphipathic helical conformation is only stable when the protein is bound to membrane.

Membrane binding by AS is reversible, due to the inherent backbone strain of the α11/3 amphipathic binding motif and the shallow insertion of α-synuclein into the membrane bilayer. This could account for the apparent transience of membrane binding by AS, and also help explain why AS does not strongly copurify with cellular membrane fractions.

This model is consistent with the proposed functions of AS in regulation of PLD activity, hDAT localization, and maintenance of presynaptic vesicle pools, although the details remain to be tested. In the case of PLD inhibition, it has been shown that inhibition by AS requires the adoption of a helical conformation. It is also notable that AS is recruited to membranes containing the product of PLD (phosphatidic acid) but not its substrate (phosphatidylcholine), an ideal situation for providing negative feedback inhibition of the enzyme.

In the case of hDAT, it appears that regulation of its activity is achieved by its translocation from the plasma membrane to an intracellular membrane compartment. Local distortion of membrane curvature by AS could influence the internalization of hDAT —containing membrane, and regulation of PLD2 activity might further regulate this endocytic event. Wersinger et al have proposed that AS tethers hDAT within the cell interior via direct interactions with the cytoskeleton.43,44 This is supported by their evidence that hDAT regulation is dependent upon cell adhesion, which supports anchoring of the cytoskeleton to the plasma membrane.

Likewise, regulation of distinct vesicle pools in the synaptic terminal could result from selective binding and tethering of specific vesicle subtypes. This is supported by EM localization of AS to an electron dense domain (presumably cytoskeletal) adjacent to presynaptic vesicles.8

This model is of course only a general framework, but it suggests potentially fruitful areas for future investigation. 1) While the available data suggest an interaction between AS and the cytoskeleton, we know almost nothing of this process. Retrograde transport of aggregated AS along microtubules has been implicated in the formation of inclusion bodies in cellular models of oxidative stress.46,47 Soluble AS is shown to bind tubulin monomers,48 but evidence for binding of AS to assembled microtubules or actin filaments under normal physiological conditions is lacking. An intriguing possibility is that AS requires a membrane surface to organize its structure and make it competent to bind the cytoskeleton. This issue deserves attention from the research community. 2) Evidence of membrane binding by AS is quite compelling, yet we cannot adequately localize this binding to particular intracellular sites due to a lack of conformation-specific probes for AS. We must devise methods to selectively detect membrane-associated AS against the background of soluble, unstructured protein. 3) It is unlikely that we have identified all of the molecular partners required for AS function. We need to develop screening techniques suitable for isolation of proteins that might interact specifically with helical AS.

Answers to these questions will be critical to understanding the breadth of activities in which AS participates. We thus conclude that further focus on membrane-dependent events should yield a much better understanding of the physiological mechanisms of α-synuclein function, both in health and disease.

Acknowledgements

M.L. Yang and J.M. George are supported by the National Institute on Aging (NIA R01 AG13762). Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the NIA.

References

1.
Maroteaux L, Campanelli JT, Scheller RH. Synuclein: A neuron-specific protein localized to the nucleus and presynaptic nerve terminal. J Neurosci. 1988;8(8):2804–2815. [PMC free article: PMC6569395] [PubMed: 3411354]
2.
Jakes R, Spillantini MG, Goedert M. Identification of two distinct synucleins from human brain. FEBS Lett. 1994;345(1s):27–32. [PubMed: 8194594]
3.
Ji H, Liu YE, Jia T. et al. Identification of a breast cancer-specific gene, BCSG1, by direct differential cDNA sequencing. Cancer Res. 1997;57(4):759–764. [PubMed: 9044857]
4.
Lavedan C, Leroy E, Dehejia A. et al. Identification, localization and characterization of the human gamma-synuclein gene. Hum Genet. 1998;103(1):106–112. [PubMed: 9737786]
5.
George JM. The synucleins. Genome Biol. 2002;3(1) [PMC free article: PMC150459] [PubMed: 11806835]
6.
George JM, Jin H, Woods WS. et al. Characterization of a novel protein regulated during the critical period for song learning in the zebra finch. Neuron. 1995;15(8):361–372. [PubMed: 7646890]
7.
Irizarry MC, Kim TW, McNamara M. et al. Characterization of the precursor protein of the nona beta component of senile plaques (NACP) in the human central nervous system. J Neuropathol Exp Neurol. 1996;55(8):889–895. [PubMed: 8759778]
8.
Clayton DF, George JM. Synucleins in synaptic plasticity and neurodegenerative disorders. J Neurosci Res. 1999;58(1):120–129. [PubMed: 10491577]
9.
Davidson WS, Jonas A, Clayton DF. et al. Stabilization of alpha-synuclein secondary structure upon binding to synthetic membranes. J Biol Chem. 1998;273(16):9443–9449. [PubMed: 9545270]
10.
Segrest J, Jones M, De Loof H. et al. The amphipathic helix in the exchangeable apolipoproteins: A review of secondary structure and function. J Lipid Res. 1992;33:141–166. [PubMed: 1569369]
11.
Jo E, McLaurin J, Yip CM. et al. Alpha-synuclein membrane interactions and lipid specificity. J Biol Chem. 2000;275(44):34328–34334. [PubMed: 10915790]
12.
Nuscher B, Kamp F, Mehnert T. et al. Alpha-synuclein has a high affinity for packing defects in a bilayer membrane: A thermodynamics study. J Biol Chem. 2004;279(21):21966–21975. [PubMed: 15028717]
13.
Perrin RJ, Woods WS, Clayton DF. et al. Exposure to long chain polyunsaturated fatty acids triggers rapid multimerization of synucleins. J Biol Chem. 2001;276(45):41958–41962. [PubMed: 11553616]
14.
Narayanan V, Scarlata S. Membrane binding and self-association of alpha-synucleins. Biochemistry. 2001;40(33):9927–9934. [PubMed: 11502187]
15.
Perrin RJ, Woods WS, Clayton DF. et al. Interaction of human alpha-synuclein and Parkinson's disease variants with phospholipids. Structural analysis using site-directed mutagenesis. J Biol Chem. 2000;275(44):34393–34398. [PubMed: 10952980]
16.
Eliezer D, Kutluay E, Bussell Jr R. et al. Conformational properties of alpha-synuclein in its free and lipid-associated states. J Mol Biol. 2001;307(4):1061–1073. [PubMed: 11286556]
17.
Bussell JrR, Eliezer D. A structural and functional role for 11-mer repeats in alpha-synuclein and other exchangeable lipid binding proteins. J Mol Biol. 2003;329(4):763–778. [PubMed: 12787676]
18.
Jao CC, Der-Sarkissian A, Chen J. et al. Structure of membrane-bound alpha-synuclein studied by site-directed spin labeling. Proc Natl Acad Sci USA. 2004;101(22):8331–8336. [PMC free article: PMC420394] [PubMed: 15155902]
19.
Palgunachari MN, Mishra VK, Lund-Katz S. et al. Only the two end helixes of eight tandem amphipathic helical domains of human apo A-I have significant lipid affinity. Implications for HDL assembly. Arterioscler Thromb Vasc Biol. 1996;16(2):328–338. [PubMed: 8620350]
20.
Kooijman EE, Chupin V, de Kruijff B. et al. Modulation of membrane curvature by phosphatidic acid and lysophosphatidic acid. Traffic. 2003;4(3):162–174. [PubMed: 12656989]
21.
Liscovitch M, Czarny M, Fiucci G. et al. Phospholipase D: Molecular and cell biology of a novel gene family. Biochem J. 2000;345(Pt 3):401–415. [PMC free article: PMC1220771] [PubMed: 10642495]
22.
Frohman MA, Sung TC, Morris AJ. Mammalian phospholipase D structure and regulation. Biochim Biophys Acta. 1999;1439(2):175–186. [PubMed: 10425394]
23.
Colley WC, Sung TC, Roll R. et al. Phospholipase D2, a distinct phospholipase D isoform with novel regulatory properties that provokes cytoskeletal reorganization. Current Biology. 1997;7:191–201. [PubMed: 9395408]
24.
Jenco JM, Rawlingson A, Daniels B. et al. Regulation of phospholipase D2: Selective inhibition of mammalian phospholipase D isoenzymes by alpha- and beta-synucleins. Biochemistry. 1998;37(14):4901–4909. [PubMed: 9538008]
25.
Ahn BH, Rhim H, Kim SY. et al. Alpha-synuclein interacts with phospholipase D isozymes and inhibits pervanadate-induced phospholipase D activation in human embryonic kidney-293 cells. J Biol Chem. 2002;277(14):12334–12342. [PubMed: 11821392]
26.
Pronin AN, Morris AJ, Surguchov A. et al. Synucleins are a novel class of substrates for G protein-coupled receptor kinases. J Biol Chem. 2000;275(34):26515–26522. [PubMed: 10852916]
27.
Payton JE, Perrin RJ, Woods WS. et al. Structural determinants of PLD2 inhibition by α-synuclein. J Mol Biol. 2004;337:1001–1009. [PubMed: 15033366]
28.
Lee C, Kang HS, Chung JK. et al. Inhibition of phospholipase D by clathrin assembly protein 3 (AP3). J Biol Chem. 1997;272(25):15986–15992. [PubMed: 9188501]
29.
Chung JK, Sekiya F, Kang HS. et al. Synaptojanin inhibition of phospholipase D activity by hydrolysis of phosphatidylinositol 4,5-bisphosphate. J Biol Chem. 1997;272(25):15980–15985. [PubMed: 9188500]
30.
Lee C, Kim SR, Chung JK. et al. Inhibition of phospholipase D by amphiphysins. J Biol Chem. 2000;275(25):18751–18758. [PubMed: 10764771]
31.
Abeliovich A, Schmitz Y, Farinas I. et al. Mice lacking alpha-synuclein display functional deficits in the nigrostriatal dopamine system. Neuron. 2000;25(1):239–252. [PubMed: 10707987]
32.
Murphy DD, Rueter SM, Trojanowski JQ. et al. Synucleins are developmentally expressed, and alpha-synuclein regulates the size of the presynaptic vesicular pool in primary hippocampal neurons. J Neurosci. 2000;20(9):3214–3220. [PMC free article: PMC6773130] [PubMed: 10777786]
33.
Cabin DE, Shimazu K, Murphy D. et al. Synaptic vesicle depletion correlates with attenuated synaptic responses to prolonged repetitive stimulation in mice lacking alpha-synuclein. J Neurosci. 2002;22(20):8797–8807. [PMC free article: PMC6757677] [PubMed: 12388586]
34.
Cole NB, Murphy DD, Grider T. et al. Lipid droplet binding and oligomerization properties of the Parkinson's disease protein alpha-synuclein. J Biol Chem. 2002;277(8):6344–6352. [PubMed: 11744721]
35.
Sharon R, Goldberg MS, Bar-Josef I. et al. Alpha-synuclein occurs in lipid-rich high molecular weight complexes, binds fatty acids, and shows homology to the fatty acid-binding proteins. Proc Natl Acad Sci USA. 2001;98(16):9110–9115. [PMC free article: PMC55381] [PubMed: 11481478]
36.
Sharon R, Bar-Joseph I, Mirick GE. et al. Altered fatty acid composition of dopaminergic neurons expressing alpha-synuclein and human brains with alpha-synucleinopathies. J Biol Chem. 2003;278(50):49874–49881. [PubMed: 14507911]
37.
Pifl C, Giros B, Caron MG. Dopamine transporter expression confers cytotoxicity to low doses of the parkinsonism-inducing neurotoxin 1-methyl-4-phenylpyridinium. J Neurosci. 1993;13(10):4246–4253. [PMC free article: PMC6576397] [PubMed: 8410185]
38.
Pifl C, Giros B, Caron MG. The dopamine transporter. The cloned target site of parkinsonism-inducing toxins and of drugs of abuse. Adv Neurol. 1996;69:235–238. [PubMed: 8615133]
39.
Lee FJ, Liu F, Pristupa ZB. et al. Direct binding and functional coupling of alpha-synuclein to the dopamine transporters accelerate dopamine-induced apoptosis. FASEB J. 2001;15(6):916–926. [PubMed: 11292651]
40.
Giasson BI, Murray IV, Trojanowski JQ. et al. A hydrophobic stretch of 12 amino acid residues in the middle of alpha-synuclein is essential for filament assembly. J Biol Chem. 2001;276(4):2380–2386. [PubMed: 11060312]
41.
Wersinger C, Sidhu A. Attenuation of dopamine transporter activity by alpha-synuclein. Neurosci Lett. 2003;340(3):189–192. [PubMed: 12672538]
42.
Wersinger C, Prou D, Vernier P. et al. Mutations in the lipid-binding domain of alpha-synuclein confer overlapping, yet distinct, functional properties in the regulation of dopamine transporter activity. Mol Cell Neurosci. 2003;24(1):91–105. [PubMed: 14550771]
43.
Wersinger C, Prou D, Vernier P. et al. Modulation of dopamine transporter function by alpha-synuclein is altered by impairment of cell adhesion and by induction of oxidative stress. FASEB J. 2003;17(14):2151–2153. [PubMed: 12958153]
44.
Wersinger C, Vernier P, Sidhu A. Trypsin disrupts the trafficking of the human dopamine transporter by alpha-synuclein and its A30P mutant. Biochemistry. 2004;43(5):1242–1253. [PubMed: 14756560]
45.
Sidhu A, Wersinger C, Vernier P. Does alpha-synuclein modulate dopaminergic synaptic content and tone at the synapse? FASEB J. 2004;18(6):637–647. [PubMed: 15054086]
46.
Lee HJ, Lee SJ. Characterization of cytoplasmic alpha-synuclein aggregates. Fibril formation is tightly linked to the inclusion-forming process in cells. J Biol Chem. 2002;277(50):48976–48983. [PubMed: 12351642]
47.
Hasegawa T, Matsuzaki M, Takeda A. et al. Accelerated alpha-synuclein aggregation after differentiation of SH-SY5Y neuroblastoma cells. Brain Res. 2004;1013(1):51–59. [PubMed: 15196967]
48.
Payton JE, Perrin RJ, Clayton DF. et al. Protein-protein interactions of alpha-synuclein in brain homogenates and transfected cells. Brain Res Mol Brain Res. 2001;95(1-2):138–145. [PubMed: 11687285]
49.
Polymeropoulos MH, Lavedan C, Leroy E. et al. Mutation in the α-synuclein gene identified in families with Parkinson's disease. Science. 1997;276:2045–2047. [PubMed: 9197268]
50.
Kruger R, Kuhn W, Muller T. et al. Ala30Pro mutation in the gene encoding alpha-synuclein in Parkinson's disease. Nat Genet. 1998;18(2):106–108. [PubMed: 9462735]
51.
Zarranz JJ, Alegre J, Gomez-Esteban JC. et al. The new mutation, E46K, of alpha-synuclein causes Parkinson and Lewy body dementia. Ann Neurol. 2004;55(2):164–173. [PubMed: 14755719]
Copyright © 2000-2013, Landes Bioscience.
Bookshelf ID: NBK6143

Views

  • PubReader
  • Print View
  • Cite this Page

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...