If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The past few years have resulted in an increased awareness and recognition of the prevalence and roles of intrinsically disordered proteins and protein regions (IDPs and IDRs, respectively) in synaptic vesicle trafficking and exocytosis and in overall synaptic organization. IDPs and IDRs constitute a class of proteins and protein regions that lack stable tertiary structure, but nevertheless retain biological function. Their significance in processes such as cell signaling is now well accepted, but their pervasiveness and importance in other areas of biology are not as widely appreciated. Here, we review the prevalence and functional roles of IDPs and IDRs associated with the release and recycling of synaptic vesicles at nerve terminals, as well as with the architecture of these terminals. We hope to promote awareness, especially among neuroscientists, of the importance of this class of proteins in these critical pathways and structures. The examples discussed illustrate some of the ways in which the structural flexibility conferred by intrinsic protein disorder can be functionally advantageous in the context of cellular trafficking and synaptic function.
A basic tenet of cell biology is that the eukaryotic cell is compartmentalized into discrete membrane-enclosed organelles and that cellular compartmentalization is critical for biological function. Trafficking between cellular compartments is achieved in part through membrane-enclosed vesicles that bud from a source membrane, travel through the cell, and fuse specifically with a given target membrane. Neurons present interesting and unique challenges for these types of processes. Lengthy axonal and dendritic processes necessitate long-distance trafficking in neurons. In addition, in contrast to the more ubiquitous and continuous nature of vesicular trafficking elsewhere in the cell, synaptic communication between neurons requires highly-localized and precisely-timed release of neurotransmitter into the synaptic cleft through fusion of the synaptic vesicle with the plasma membrane. This process is highly regulated by a number of accessory proteins (
), some of which are intrinsically disordered or contain functionally critical disordered regions. Accordingly, intrinsically disordered proteins and regions play critical roles in synaptic transmission.
Protein structure exists as a continuum, with folded, ordered, and well-structured proteins and domains at one extreme and flexible, dynamic, and intrinsically disordered proteins and regions (IDPs
) for a discussion of the term “disordered” versus “unstructured”). This review begins with a brief summary of biophysical properties of IDPs and their interactions before discussing a number of notable examples of how intrinsic disorder contributes to vesicular trafficking in general. We then focus on structural disorder within a number of proteins critical for synaptic function, especially synaptic vesicle release and recycling. Finally, we close with a discussion of the potential role of IDP/IDR-mediated phase transitions and membrane-less organelles in the organization of key elements of the synapse.
Unique IDP properties confer conformational and functional flexibility
The primary sequences of IDPs contain a high proportion of charged residues, with few hydrophobic amino acids (
). Although IDPs feature relatively simple primary sequences, their inability to spontaneously fold into a unique three-dimensional structure leads to great structural complexity. Charge content and patterning within IDP sequences alter the extent of chain collapse, and the sequence composition determines how IDPs respond to external factors like ionic strength and temperature (
Conformational flexibility allows IDPs to interact with other macromolecules in a variety of ways. Indeed, IDPs can be promiscuous binders capable of interacting not only with multiple proteins (which may be other IDP/IDRs or structured proteins/domains), but also with lipid membranes or nucleic acids. IDP interactions often involve folding of the IDP/IDR, but folding upon binding is not an absolute requirement (
). In conformational selection, some subset of the IDP structural ensemble adopts a conformation appropriate for binding, and the partner subsequently interacts with this preformed structure. In induced fit, binding precedes folding via an initial encounter complex (
). IDP/protein interactions tend to be of low affinity yet high specificity, a feature that is often coupled to regulatory functions within signaling networks: the interactions can be easily and rapidly turned “on” or “off” as required (
). Indeed, an unfolded peptide chain is typically more accessible to modifying enzymes. PTMs change the physicochemical properties of the primary sequence; this produces a variety of structural changes, which then leads to alteration and expansion of IDP function. Specifically, PTMs can alter a given protein's steric, hydrophobic, or electrostatic properties, can stabilize, destabilize, or induce local structure, and can inhibit or enhance long-range tertiary contacts. PTMs alter the energy landscape and resultant conformational ensemble of the IDP, and they modulate interactions with other biomolecules (
The trafficking of synaptic vesicles is a specialized case of cellular trafficking, which in general requires that vesicles carrying the appropriate cargo bud from a source membrane, travel in the appropriate direction, and then fuse with the proper cellular target. It is therefore relevant to examine the ways in which structural disorder may contribute to general vesicle trafficking pathways before examining contexts that are more specific to trafficking at the synapse. Using primary sequence analysis, Pietrosemoli et al. (
) examined protein disorder in cellular trafficking pathways such as clathrin-mediated endocytosis and transport between the ER and Golgi mediated by COPI (Golgi to ER) and COPII (ER to Golgi) coat proteins. They found that proteins associated with enzymatic activity and proteins that function as adaptors for vesicle cargo featured especially high degrees of disorder.
Long disordered regions mediating protein/protein interactions were often found adjacent to structured catalytic domains (
). Synaptojanins, for example, contain three distinct domains: an N-terminal inositol phosphatase domain; a central inositol phosphatase domain; and a disordered C-terminal domain featuring a proline-rich domain and three asparagine–proline–phenylalanine (NPF) repeats (Fig. 1A). The C-terminal domain interacts with other proteins involved in endocytosis, including the AP2 adaptor complex, amphiphysin, endophilin, DAP160/intersectin, syndaptin, and Eps15 (
). Importantly, synaptojanin-1 is a synapse-specific family member that plays a critical role in the endocytosis of synaptic vesicles and is also mutated in some forms of familial Parkinson's disease (
As another example, the GTPase-activating protein (GAP) ArfGAP1, which regulates the enzymatic activity of ADP-ribosylation factor 1 (Arf1), contains disordered amphipathic lipid-packing sensor (ALPS) motifs that bind to membranes and sense membrane curvature to contribute critically to COPI trafficking (
). The ArfGAP1 ALPS motifs bind avidly to small liposomes in an α-helical conformation, with membrane binding predominantly driven through insertion of a number of bulky hydrophobic residues on the hydrophobic face of the amphipathic helices (
). This suggests a feedback loop (Fig. 1B): Arf1/COPI first induce membrane deformation and so enhance membrane curvature, which in turn stimulates ArfGAP1 activity, enhances Arf1 GTP hydrolysis, and leads to coat disassembly (
). Although COPI vesicles are not present at the synapse, ArfGAP1 provides an example of how IDP/membrane interactions contribute to vesicle trafficking. Additional synapse-specific examples are discussed further below.
An example of a disorder-containing adaptor protein is the Epsin family of proteins. Epsins contain a folded epsin N-terminal homology domain at their N terminus that interacts with membranes, whereas the rest of the protein is disordered (
), with a protein-binding site at one end of a disordered chain. The larger resulting hydrodynamic radius allows a binding module to explore a greater volume and thereby helps increase association rates (
). Like synaptojanin-1, epsin-1 is thought to play a role in clathrin-mediated synaptic vesicle endocytosis/recycling.
Intrinsically disordered SNARE proteins as the core membrane fusion machinery
The SNARE proteins (for SNAP receptor proteins, where SNAPs are soluble NSF attachment proteins, and NSF is the N-ethylmaleimide-sensitive factor) are a superfamily of small proteins that mediate membrane fusion in all steps of cellular secretory pathways and that constitute the core membrane fusion machinery (
). Humans contain 36 different SNARE proteins, which are found attached to membranes, often through a C-terminal transmembrane domain, although attachment via post-translational lipidation also occurs for some family members. SNAREs are often classified as v- or t-SNAREs depending on their localization to vesicles or to the target membrane, respectively. Membrane fusion events obviously are critical for synaptic vesicle exocytosis, where the most relevant SNAREs are the v-SNARE synaptobrevin-2 and the t-SNAREs syntaxin-1 and synaptosome-associated protein of 25 kDa (SNAP-25).
The SNARE motif is a conserved stretch of 60–70 amino acids featuring heptad repeats (
). SNARE motifs localized to two closely apposed membranes can zipper in an N- to C-terminal direction into a parallel four-helix coiled coil termed the SNARE complex. The coiled-coil interface consists of 16 stacked layers of interacting side chains, mostly featuring hydrophobic interactions, with a central “0” layer that contains three highly-conserved glutamine residues and one highly-conserved arginine residue (SNAREs may also be classified as Qa-, Qb-, Qc-, and R-SNAREs depending on the identity of this residue) (
). The free energy released through the SNARE assembly process is thought to drive fusion of the two membranes, although precisely how the energetics of SNARE assembly are coupled to membrane fusion remains unclear. In one model, SNARE complex folding may subsequently propagate through the transmembrane domain to exert a mechanical force on fusing membranes (illustrated using the synaptic SNARE proteins in Fig. 2); this has yet to be definitively established, however (
). Regardless, SNAREs represent a clear example of how coupled binding and folding by a disordered region fulfills a critical cellular function. After assembly, the AAA ATPase NSF, along with SNAP proteins, disassembles post-fusion SNARE complexes to “recharge” the system with the necessary energy required for future rounds of fusion (
). In addition, as noted above, SNARE assembly occurs through N- to C-terminal zippering, and this process can be frozen at intermediate stages by protein-binding partners (e.g. complexin, discussed below). Such intermediate species may include partially zippered SNARE complexes, binary and/or ternary SNARE complexes, or complexes containing SNARE protein(s) along with one or more non-SNARE–binding partners (
). Indeed, for proper assembly, SNAREs must interact with a number of other proteins, and in many cases these interactions involve some degree of SNARE folding. Thus, their structural flexibility likely allows the SNARE proteins to engage a variety of partners through discrete bound state conformations and in various multimeric states, a feat that is critical for their proper regulation and function.
Structural disorder within accessory proteins that regulate synaptic vesicle exocytosis
The three synaptic SNARE proteins alone are sufficient to drive some degree of membrane fusion in in vitro fusion assays. In vivo, however, their behavior is tightly regulated and modulated by a number of accessory proteins, including complexin, synaptotagmin (the calcium sensor that triggers action potential evoked exocytosis by coupling calcium influx to SNARE assembly), Munc18, Munc13, tomosyn, RIM proteins, RIM-binding proteins (RIM-BPs), Rab proteins, ELKS, α-Liprins, and Piccolo/Bassoon proteins (
). This extensive machinery utilizes abundant protein/protein and protein/membrane interactions to structurally and functionally organize active zones, specialized electron-dense presynaptic membrane-associated sites of synaptic vesicle exocytosis that appose post-synaptic densities (the reader is referred to Refs.
for further information regarding active-zone proteins, composition, organization, and architecture). At active zones, synaptic vesicles are docked and primed adjacent to the requisite calcium channels so that evoked fusion can occur efficiently when required, with a tight and rapid (sub-millisecond) coupling between the influx of calcium that results from the arrival of an action potential and vesicle release (
). SNAREs can also mediate low levels of spontaneous synaptic vesicle fusion with the plasma membrane, but for optimal action-potential–based inter-neuronal signaling, such spontaneous exocytosis must be kept to a minimum by regulatory proteins (
). Importantly, intrinsically disordered regions from a number of SNARE accessory/regulatory proteins play key functional roles within this extensive interaction network to help organize the active zone and optimize the synapse for rapid exocytosis.
Tomosyn is a SNARE-binding presynaptic protein that is considered to be an inhibitor of synaptic vesicle release, possibly at the priming step. Tomosyn contains two conserved domains, including a C-terminal region with homology to the R-SNAREs, and an N-terminal WD40 repeat region (Fig. 3A). As with other SNAREs, the tomosyn SNARE motif is disordered in the free state, but it can dimerize with syntaxin-1 (
). Through ternary SNARE complex formation, tomosyn can regulate synaptobrevin-2 binding to syntaxin-1 and SNAP-25, and this R-SNARE substitution mechanism may underlie the inhibition of exocytosis by tomosyn overexpression in PC12 cells (
). Alternatively, tomosyn binding to the SNARE motif of syntaxin-1 may displace Munc18 from syntaxin; it was proposed that tomosyn may activate syntaxin-1 and allow it to interact with synaptobrevin-2 (
Synaptotagmin-1 (Syt1) has been identified as a key calcium sensor linking a presynaptic membrane depolarization-induced calcium influx to evoked exocytosis. Syt1 contains an N-terminal transmembrane α-helix followed by two C2 domains, C2A and C2B, which bind three and two Ca2+ ions, respectively, through calcium-binding loops that can then interact with membranes. Syt1 thus interacts with phospholipid membranes in a Ca2+-dependent manner. Syt1 also interacts with SNAREs and can bind simultaneously to membranes and membrane-anchored SNARE complexes to form the so-called quaternary SNARE–synaptotagmin-1–Ca2+–phospholipid (SSCAP) complex (
). This complex could facilitate fusion by inducing negative membrane curvature, with the Syt1/membrane interaction serving to bring apposed membranes together. Alternatively, Syt1 may induce positive curvature in the plasma membrane. Interestingly, and similarly to complexin, Syt1 (and Syt2) also normally clamp spontaneous SV release, and the activating versus inhibitory functions of synaptotagmin are likely independent (
An intrinsically disordered region between the synaptotagmin transmembrane domain and C2A has been shown to be essential for both calcium-independent vesicle docking, and calcium-dependent fusion pore opening (
). Interestingly, this IDR contains an N-terminal part rich in basic amino acids and a C-terminal part rich in acidic amino acids. This segregation of charge seems to result in a molecular zipper that contributes to fusion pore opening when closed but that facilitates vesicle docking when open (Fig. 3B); shortening the linker or cross-linking it into a folded conformation reduces docking but enhances fusion pore opening. These observations led to a model whereby this disordered linker region extends to facilitate vesicle docking (perhaps through a sort of fly-casting mechanism) but then folds in a way that facilitates fusion pore opening (
), it appears to be weakly stable. It was proposed that this marginal stability allows C2A to adopt and fluctuate between multiple conformational states with unique functions and behaviors, which endow synaptotagmin with great functional diversity (
). 1) It acts as a chaperone for syntaxin-1 that facilitates proper syntaxin-1 localization and expression. Indeed, in contrast to its apparently essential facilitatory role in synaptic vesicle fusion, Munc18 was originally shown to bind to the “closed” conformation of syntaxin in a way that would be expected to inhibit membrane fusion (with the N-terminal Habc domain of syntaxin folded back onto the SNARE domain). In this conformation, Munc18 stabilizes syntaxin-1 and facilitates transport of syntaxin-1 to the plasma membrane. 2) Munc18 facilitates priming via promotion of SNARE-mediated membrane fusion in several different ways. Munc18 was later shown to bind to the syntaxin N-peptide (at the very N terminus) in a conformation that is compatible with SNARE complex assembly. It also binds syntaxin–1-SNAP-25 heterodimers and could thus facilitate or nucleate SNARE complex assembly. In addition to binding t-SNAREs like syntaxin, it can also simultaneously bind to v-SNAREs, possibly aligning them so as to facilitate the initial steps of SNARE complex formation (
). Furthermore it may enable lipid mixing either directly or through lipid destabilization, and it may help to spatially and asymmetrically organize assembled SNARE complexes around the fusion site by preventing diffusion of the bulky Munc18–SNARE complex assembly to the center of the synaptic vesicle/plasma membrane intermembrane space. 3) Finally, it has been suggested that Munc18 contributes to docking of large dense-core vesicles (
). Interestingly, in the Munc18–syntaxin-1 complex, two anti-parallel helices in domain 3a, helices 11 and 12, are connected by a 21-residue bent hairpin loop with an irregular conformation. Eight residues of this loop are disordered as they do not appear in crystal structures, and the remaining visible residues have high B-factors, indicating increased mobility. This domain 3a loop is essential for exocytosis from PC12 cells but is not required for syntaxin-1A transport (
). The disordered residues in the loop connecting helices 11 and 12 thus appear to contribute to the structural flexibility necessary for Munc18 to pivot between the requisite structures for its inhibitory and facilitatory functions (Fig. 3C).
Mouse synapse-defective 1A
Synapse–defective-1 proteins, including mSYD1A (mouse synapse–defective-1A), regulate presynaptic differentiation. Many SYD-1 proteins contain a RhoGAP domain and a PDZ domain that links SYD-1 to the surface receptor neurexin (a presynaptic protein that helps connect neurons across the synaptic cleft); these domains may mediate SYD-1 function. Syd1 mutations in Caenorhabditis elegans perturb proper localization of active-zone components and synaptic vesicles (
). The first mammalian SYD-1 ortholog was recently identified in mice (mSYD1A) and shown to contribute to vesicle docking, insofar as mSYD1A knockout hippocampal synapses contain fewer docked vesicles and reduced synaptic transmission. Interestingly, mSYD1 function depends on a multifunctional intrinsically disordered domain (Fig. 3D). In contrast to invertebrate SYD-1, mSYD1A lacks the PDZ domain and contains an active RhoGAP domain (which is inactive in invertebrates). mSYD1A RhoGAP activity is regulated through intramolecular interactions that include the disordered domain. Furthermore, the disordered domain acts as an interaction module capable of interacting with multiple proteins, including Munc18-1 as well as a LAR–LiprinA2 complex. The protein/protein interactions mediated by this disordered domain likely contribute to organization of the active zone and to proper vesicle docking (
The examples above illustrate both the importance of disordered protein regions in the regulation of SNARE-mediated synaptic vesicle exocytosis and some of the mechanisms involved. Disordered regions can compete with other disordered binding motifs for binding partners, can served as a platform for recruiting multiple interaction partners to provide scaffolding/organizing functions, and can undergo highly-specific or less well-defined conformational transitions that regulate function and activity. The diversity of their modes of action contributes to the ubiquitous involvement of disordered proteins in various biological processes, but also constitutes a challenge in uncovering and understanding their functional roles.
Protein/lipid interactions at the synapse: functional significance of membrane binding by IDPs
A number of IDPs interact with lipid membranes. This interaction can be driven by partitioning of hydrophobic residues into the membrane as well as by electrostatic interactions between charged protein residues and lipid headgroups (
). Some IDPs instead remain unfolded in the membrane-bound state, as is the case, for example, with the MARCKS peptide (a membrane-binding region of the myristoylated alanine-rich protein kinase C substrate, or MARCKS, protein, which can sequester phosphatidylinositol 4,5-bisphosphate (PIP2) in the membrane and regulate phospholipase C signaling) (
). A number of intrinsically disordered membrane-binding proteins function as membrane curvature sensors and membrane curvature generators. Above, ArfGAP1 was cited as an example of how membrane curvature-selective binding contributes to IDP function in the context of vesicle trafficking. This type of behavior is observed at the synapse as well, as will be discussed for two intrinsically disordered pre-synaptic membrane curvature sensors, α-synuclein and complexin.
Complexins form a family of small, highly-charged, and highly-conserved cytoplasmic proteins that have emerged as key regulators of SNARE-mediated synaptic vesicle exocytosis (
). Although complexins are intrinsically disordered in their free state, they can directly interact with assembled and/or assembling SNARE complexes through a central α-helical domain that binds in an antiparallel fashion in the groove formed between syntaxin-1 and synaptobrevin-2 (
). Complexins feature four different domains with discrete functions in synaptic vesicle exocytosis, including a facilitatory N-terminal domain (NTD), an inhibitory accessory helix/accessory domain (AH), the above-mentioned essential central helix (CH), and a C-terminal domain (CTD) that can bind lipid membranes and that may inhibit or facilitate exocytosis (
), often promote or facilitate the binding of IDPs/IDRs to their interaction partners.
Complexin expression levels are altered in a variety of neurological and psychiatric disorders. Although such expression level changes may not be causal in these disorders, they could contribute to their corresponding symptomatology (
). Complexin appears to either promote or inhibit vesicle fusion depending on the experimental approach used; complexin also appears to differentially affect action potential-evoked versus spontaneous vesicle exocytosis and to do so in a species-dependent fashion. Overall, a general consensus has emerged that complexins inhibit spontaneous release while facilitating synchronous, evoked release (
). The detailed mechanisms by which complexin fulfills these distinct functions, however, remain unclear. Regardless, complexin's inhibitory and facilitatory functions appear to be distinct and separable and so likely operate through discrete mechanisms (
). As noted above, the four complexin domains likely have discrete functions in vesicle exocytosis, and complexin domain function has been extensively dissected. Here, we focus on the highly-disordered NTD and CTD, while noting the importance of the structured CH and AH domains.
The complexin NTD has been reported to facilitate vesicle release (
). It is interesting to note that the structures of the SNARE bundle-bound complexin CH suggest that the NTD and AH domains will be situated in the region where trans-SNARE complexes insert into the membrane. In mouse neurons, mutation of the membrane insertion sequence of synaptobrevin-2 generated a similar phenotype to that of complexin knockout as assessed by electrophysiology. Together, these observations led to the hypothesis that although these two N-terminal regions of complexin fulfill discrete facilitatory (NTD) and inhibitory (AH) functions, together they control the force transfer from SNARE complexes to membranes during fusion (
). It is likely that complexin interacts with SNAREs in a variety of ways and that these varied interactions contribute to the dual functions of complexin. A recent study provided evidence for both cis- and trans-conformations of the complexin–SNARE complex. The cis-conformation, which may help to activate synchronous neurotransmitter release, required the N-terminal domain of complexin in addition to the accessory and central helices (
). Although disordered when free in solution, the first ~17 residues of the complexin N-terminal domain appear somewhat conserved across species and, when modeled as a helix, show potentially amphipathic helical character. Methionine 5 and lysine 6 within this region are particularly critical for the facilitation of evoked release (
). Instead, it was hypothesized that the NTD might bind to the C terminus of the SNARE complex (specifically between SNAP-25 and syntaxin-1) and, in so doing, either stabilize the SNARE complex and/or release the inhibitory function of complexin (
). Subsequent studies did, however, show a potential interaction between the N-terminal amphipathic region and lipid membranes; interestingly, NTD function did not require it to be covalently attached to the rest of complexin, and it could be functionally substituted by an unrelated hemagglutinin fusion peptide (
). Notably, the aforementioned homozygous mutations to complexin observed in patients with severe infantile myoclonic epilepsy and intellectual disability could potentially perturb complexin/membrane interactions, especially given that one is a C-terminal truncation mutation. The impact of these mutations has not yet been formally assessed, however.
The CTDs of both mouse and worm complexin have been established by NMR as intrinsically disordered (
). The CTDs of worm and mammalian complexin have been shown to interact with liposomes through a conserved amphipathic region, which in the worm protein corresponds to the location of residual helicity in the unbound state (
). The CTD/membrane interaction may contribute to inhibition and/or facilitation of synaptic vesicle fusion by the CTD. We previously established that the worm complexin CTD contains tandem lipid-binding motifs that together sense membrane curvature and selectively bind to more highly-curved membranes (
). A C-terminal motif rich in bulky hydrophobic residues and positively charged lysine residues initiates binding and likely remains unstructured in the membrane-bound state; an adjacent amphipathic motif adopts helical structure upon membrane binding, but only for highly-curved membrane surfaces (
). The structure of membrane-bound mouse complexin has not, however, been as extensively characterized. The exact mechanisms by which membrane binding might contribute to function remain unclear. We and others have proposed that the CTD localizes complexin to synaptic vesicles (Fig. 4A), an interaction that is required for inhibition of spontaneous exocytosis by complexin; this interaction would be facilitated by the preferential binding of complexin to the highly-curved synaptic vesicle membrane (
). This suggests a role for the CTD beyond membrane binding. It may, for example, mediate complexin protein/protein interactions, an idea supported by a recent report of potential CTD/SNARE interactions (
Finally, the impact that CTD PTMs may have on membrane association and/or protein interactions remains essentially uncharacterized. The CTD contains potentially phosphorylatable serine and threonine residues, and mutation of human complexin serine 115–one putative site for such phosphorylation–impaired the ability of complexin to stimulate liposome fusion in an in vitro assay, although membrane binding by these mutants was not examined (
). In flies, protein kinase A (PKA)-dependent phosphorylation of (fly complexin) serine 126 is required for synaptic plasticity and growth. Phosphorylation of this residue occurs as a result of retrograde signaling following stimulation, and it impairs the ability of complexin to clamp spontaneous fusion. The resultant enhanced spontaneous release then contributes to activity-dependent synaptic growth (
Ultimately, a complete understanding of the function of the complexin CTD remains elusive, and the detailed mechanisms by which the CTD and CTD/membrane interactions exert their function remain unclear. Nonetheless, it provides a key example of how an intrinsically disordered region mediates a selective and functionally requisite interaction with the synaptic vesicle membrane. Future work will be required to tease out further details, including how PTMs alter CTD function and interactions. Overall, despite its small size, complexin features at least two distinct activities that are modulated in multiple ways by multiple regions, namely SNARE and synaptic vesicle binding allow complexin to clamp or facilitate exocytosis. Its intrinsically disordered character may play a key role in providing the flexibility required for carrying out opposing functions in different contexts utilizing different structures and interactions.
α-Synuclein is a small, soluble, and predominantly presynaptic protein that has been implicated in Parkinson's disease as well as other neurodegenerative “synucleinopathies.” Mutations in α-synuclein have been linked to familial forms of these diseases (
). The N-terminal domain of α-synuclein has been clearly established as a membrane curvature sensor that folds into an amphipathic helix upon membrane binding and that displays enhanced binding to more highly-curved vesicles; binding is further enhanced by negatively charged lipids and by conical lipids (e.g. lipids with phosphatidylethanolamine headgroups) (
). Interestingly, synuclein can, in some contexts, actively induce membrane curvature to remodel lipid membranes; this could have functional significance for synaptic vesicle exocytosis, which requires alterations in membrane shape and curvature prior to membrane fusion (
). Its N-terminal ~100 residues constitute a lipid-binding domain (Fig. 5A) featuring seven imperfect 11-residue repeats that can adopt an amphipathic helical structure upon binding to detergent micelles or phospholipid vesicles (
Once membrane binding is initiated at the N-terminal end of α-synuclein, coupled membrane binding and folding propagates the initial helical structure through the remainder of the N-terminal lipid-binding domain. Several different membrane-bound helical conformations have been reported featuring amphipathic helices of varying lengths that lie along the surface of the membrane. In the extended-helix conformation, the entire N-terminal domain (~100 residues) binds to the membrane surface through a continuous amphipathic α-helix with an unusual 11/3 periodicity (wherein 11 residues form 3 helical turns, a slightly overwound geometry compared with α-helices, where 18 residues form 5 turns) (