Skip to main content
NCBI home page
Protein Science : A Publication of the Protein Society logoLink to Protein Science : A Publication of the Protein Society
. 2018 Jul 5;27(7):1314–1324. doi: 10.1002/pro.3426

Structure and dynamics of the extended‐helix state of alpha‐synuclein: Intrinsic lability of the linker region

Yoon‐Hui Sung 1, David Eliezer 1,✉
PMCID: PMC6032355  PMID: 29663556

Abstract

The Parkinson's protein alpha‐synuclein binds to synaptic vesicles in vivo and adopts a highly extended helical conformation when binding to lipid vesicles in vitro. High‐resolution structural analysis of alpha‐synuclein bound to small lipid or detergent micelles revealed two helices connected by a non‐helical linker, but corresponding studies of the vesicle‐bound extended‐helix state are hampered by the size and heterogeneity of the protein‐vesicle complex. Here we employ fluorinated alcohols (FAs) to induce a highly helical aggregation‐resistant state of alpha‐synuclein in solution that resembles the vesicle‐bound extended‐helix state but is amenable to characterization using high‐resolution solution‐state NMR. Analysis of chemical shift, NOE, coupling constant, PRE and relaxation measurements shows that the lipid‐binding domain of alpha‐synuclein in FA solutions indeed adopts a single continuous helix and that the ends of this helix do not come into detectable proximity to each other. The helix is well ordered in the center, but features an increase in fast internal motions suggestive of helix fraying approaching the termini. The central region of the helix exhibits slower time scale motions that likely result from flexing of the highly anisotropic structure. Importantly, weak or missing short‐ and intermediate‐range NOEs in the region corresponding to the non‐helical linker of micelle‐bound alpha‐synuclein indicate that the helical structure in this region of the protein is intrinsically unstable. This suggests that conversion of alpha‐synuclein from the extended‐helix to the broken‐helix state represents a functionally relevant structural transition.

Keywords: alpha‐synuclein, Parkinson, extended‐helix, membrane, HFIP, aggregation, amyloid

Introduction

Alpha‐synuclein (aS) is a small highly conserved vertebrate protein with presynaptic localization that is linked to Parkinson's disease (PD) by genetics and pathology. aS binds to membranes such as axonal transport vesicles,1 lipid rafts,2 lipid droplets,3 and outer membranes of Saccharomyces cerevisiae.4 In vitro studies have shown that aS binds to negatively charged small unilamellar vesicles (SUVs) and detergent micelles.5, 6, 7, 8, 9, 10, 11 The N‐terminal lipid‐binding domain of aS is composed of residues 1–1021 1 and contains seven imperfect 11 residue repeats (Figure 1). When associated with SUVs or micelles, this domain undergoes conformational changes from a highly disordered state11, 12, 13, 14 to highly helical structure that nevertheless lacks any stable tertiary interactions. In contrast, the highly acidic C‐terminal region remains largely unbound and unstructured in the presence of micelles or vesicles.5, 6, 7, 8, 9, 11, 15, 16 NMR and ESR studies of aS associated with negatively charged micelles revealed the formation of two long helices spanning residues V3‐V37 and K45‐T92, connected by a non‐helical linker that interacts with the membrane surface.6, 7, 9, 16 In contrast, upon binding to SUVs, a long single helix is formed, as indicated by ESR and FRET measurements of spin‐labeled or fluorescently labeled aS associated with SUVs.17, 18, 19, 20 Interestingly, the association of aS with SUVs in vitro10, 21 and with membranes in vivo22, 23 appears to be weak, whereas association with micelles,11, 24 or with vesicles still attached to synaptosomal membranes25 is tight. This has lead us to propose a model in which the extended‐helix state of vesicle‐bound aS serves as a precursor to a more tightly bound state in which the broken‐helix conformation enables the protein to bridge between two closely apposed membranes. For example, we posit that aS could bridge between the vesicle and plasma membranes of a docked synaptic vesicle.17, 26, 27 The idea that aS may be able to bridge different membranes has been supported in other recent studies as well.28

Figure 1.

Figure 1

Open in a new tab

Sequence of human aS. The imperfect 11‐residue repeats are delineated by thin horizontal lines and exons are indicated as thick horizontal lines. The positions of three mutations associated with familial Parkinson's disease are indicated by boldface characters.

While the broken helix state of aS has been elucidated in great detail using NMR spectroscopy of the micelle‐bound protein, the extended‐helix state has defied direct observation via NMR because of severe resonance broadening,11, 21, 29 even in the context of small lipid bilayer constructs such as nanodiscs.30 Here we take advantage of the fact that fluorinated alcohols (FAs) such as trifluoroethanol (TFE) and hexafluoroisopropanol (HFIP) can be used to investigate the effects of apolar environments on the conformations of proteins and peptides. High concentrations of FAs have been shown to induce helical conformations preferentially in protein regions with an intrinsic helical tendency. Although the mechanisms by which FAs induce helical conformations remain controversial, these effects have been partly explained by the stabilization of exposed hydrophobic residues by a hydrophobic microenvironment provided by FA hydrophobic clusters in water/alcohol mixture.31, 32, 33 Interestingly, low concentrations of FAs have been shown to enhance the formation of β‐sheet rich aggregates or fibrils by aS, as well as other amyloidogenic proteins such as the Alzheimer's protein tau,34 whereas higher concentrations are protective against aggregation.35, 36, 37, 38 These effects are similar to those observed when aS is in the presence of low concentrations (enhanced aggregation) or of high concentrations (protection from aggregation) of lipid membranes,27 supporting the idea that FAs may faithfully recapitulate the effects of lipid membranes on the conformations and behavior of aS. Based on this observation, and because the membrane‐bound extended‐helix state of aS has proven recalcitrant to high resolution characterization, either by NMR or other methods, we set out here to characterize the structural and dynamical properties of aS at high concentrations of FAs and to compare them with those observed for the micelle‐bound broken helix state of the protein to provide a deeper understanding of the helical conformations that aS samples when exposed to hydrophobic environments.

Results

The lipid‐binding domain of aS adopts highly helical structure in 30% hexafluoroisopropanol (HFIP)

Prior studies of aS showed that TFE or HFIP at concentrations above ∼30% by volume lead to the formation of a highly helical state of aS that is protected from aggregation,35, 36, 38 suggesting that these conditions produce a state corresponding to the extended‐helix state observed for liposome‐bound aS. FA‐water mixtures are quite viscous, and attempts to obtain high quality NMR spectra of aS in 30% TFE were not successful. As an alternative, we considered the FA HFIP, which has a somewhat lower viscosity than TFE. The CD spectrum of C‐terminally truncated aS (residues 1–102, corresponding to the lipid‐binding domain of aS and to exons 1–3 of the SNCA gene, hereafter referred to as aS103stop) in 30% HFIP was measured to assess the overall secondary structure content, and compared to a corresponding spectrum obtained for the SDS micelle bound state of the protein (Fig. 2). The spectra indicate similarly high helical propensities in both conditions with characteristic double negative minima at 208 and 222 nm, and a positive maximum around 195 nm. The data suggest a slightly higher degree of helical propensity in 30% HFIP than in the presence of SDS micelles, as indicated by the greater signal intensities at 195, 208, and 222 nm, although difference is small and could also result from experimental error in the protein concentrations.

Figure 2.

Figure 2

Open in a new tab

Far UV CD spectra of aS103stop in the presence of 40 mM SDS and 30% HFIP at pH 7.4.

aS in 30% HFIP gives rise to tractable 2D NMR spectra

The 1H‐15N heteronuclear single quantum coherence (HSQC) spectrum of aS103stop in 30% HFIP (Fig. 3) exhibits well resolved resonances, with a limited dispersion in the proton dimension (ca. 1.4 PPM) indicating a lack of tertiary structure, that is nevertheless greater than that observed for the free state of aS (ca. 1 PPM).11 This is consistent with the presence of helical secondary structure, as indicated by the CD data. To enable site‐specific analysis of structural elements of aS in 30% HFIP, the backbone and Cβ NMR resonances were assigned (see Materials and Methods). All NMR measurements were performed at a temperature of 323 K to increase the tumbling rate of the protein, which is slowed by the higher viscosity of HFIP compared with water. Sequence specific resonance assignments are annotated in Figure 3.

Figure 3.

Figure 3

Open in a new tab

1H‐15N heteronuclear single quantum coherence (HSQC) spectra of aS103stop in the presence of 30% HFIP at pH 7.4, at 50°C. Sequence‐specific resonance assignments are indicated. Unlabeled peaks correspond to side‐chain NH2 groups.

Chemical shifts indicate that 30% HFIP induces an uninterrupted helical structure throughout the lipid‐binding domain of aS

To identify the type and location of secondary structure, the chemical shifts of the assigned Cα resonances for aS103stop in 30% HFIP were compared to the values that would be expected for a random coil ensemble of conformations.39 These secondary shifts, defined as the difference between the observed and random coil chemical shifts, have a strong correlation with protein secondary structure,40, 41 with helical structure indicated by positive Cα secondary shifts, while extended or strand‐like structure is indicated by the negative secondary shifts. Unstructured flexible backbone regions exhibit secondary shifts closer to zero values. Figure 4 shows the Cα secondary chemical shifts of as a function of residue number for aS103stop in 30% HFIP, compared with the corresponding values for SDS micelle‐bound aS103stop.21 The overall patterns of the secondary chemical shifts of aS103stop in both conditions indicate extensive helical structure (secondary shifts ca. 3 PPM or greater) throughout the lipid‐binding domain, consistent with the CD data. Nevertheless, some regions exhibit increased helical propensities in 30% HFIP compared with SDS. The most noticeable difference between the two conditions appears in the region spanning residues 37–44. In particular, in the SDS micelle‐bound state two very low (< 1 PPM) secondary shifts at residues 43 and 44 indicate a break in the helical structure, whereas in HFIP these secondary shifts are considerably larger (2–4 PPM). Furthermore, secondary shifts for residues 37–40 are in the range of 1–2 PPM in the micelle‐bound state compared to 4–5 PPM in HFIP. These observations suggest that the interruption in helical structure observed in this region for the micelle‐bound state is absent in the conformation adopted by the protein in 30% HFIP, resulting in a long uninterrupted helical structure. In addition, other regions near to where the helical structure of micelle‐bound aS has been noted to be more dynamic and less regular,6, 7, 9 such as residues 28–36 and residues 62–66, feature somewhat higher secondary shifts in HFIP.

Figure 4.

Figure 4

Open in a new tab

Comparison of the deviation of Cα chemical shifts from random coil values (so called secondary shifts) for aS103stop in the presence of 30% HFIP (bar) and 40 mM SDS (red line). Cα values for the SDS‐bound states are taken from previous reports.21, 66.

NOEs and coupling constants support formation of a single extended aS helix in 30% HFIP

In addition to chemical shifts, we examined short and medium range Nuclear Overhauser Effects (NOEs) as well as the vicinal coupling constant, 3JHNHA for aS103stop in 30% HFIP, as summarized in Figure 5. Since the distance between sequential amide protons is short in helical conformations and considerably longer in extended or strand structure, the NOEs between these nuclei are relatively stronger in helical conformations than in extended or strand structure. Strong NOEs between sequential amide protons are observed throughout the sequence of aS103stop in 30% HFIP and medium range NOEs, including NH‐NH(i,i + 2) and Hα‐NH(i, i + 3) are detected for most residues, consistent with a single uninterrupted helical stretch, as also indicated by the chemical shift analysis. Nevertheless, sequential amide proton NOE intensities were weaker for residues 41–45 and NH‐NH(i,i + 2) NOEs are largely missing in this area. Because these NOEs are typically weaker in more dynamic regions of polypeptides, this region, which forms the non‐helical linker region in the micelle‐bound state, likely retains some flexibility in 30% HFIP despite populating a helical structure.

Figure 5.

Figure 5

Open in a new tab

Summary sequential and medium range NOE connectivities and coupling constants observed in aS103stop in the presence of 30% HFIP. The size of the bar corresponds to the intensity of the NOE. Open bars indicate resonance overlap which precludes unambiguous assignment. The vicinal coupling constants, 3JHNHααdetermined for aS103stop in 30% HFIP are displayed as filled and open circles corresponding to 3JHNHααcoupling constants <6.0 Hz or >6.0 Hz, respectively.

Vicinal coupling constants, 3JHNHA are also useful for assessing the secondary structure of proteins. In helical structures, 3JHNHA values of 4–5 Hz are expected while extended structures have 3JHNHA in the range of 8–9 Hz.42, 43 For example, the conformational dependence of 3JHNHA coupling constants estimated from a data base of high resolution crystal structures resulted in 3JHNHA values of 4.8, 5.6, and 8.5 Hz for alpha‐helix, 310‐helix and beta‐strand, respectively.44 For less rigid polypeptides, the measured 3JHNHAα coupling constant is a population weighted average over the distribution of angles sampled by a given conformational ensemble.45 In such cases, 3JHNHA constants less than 6 Hz suggest presence of helical structure. All measurable coupling constants for aS103stop in 30% HFIP are in the range of 4–6 Hz throughout the sequence except for F4 (6.12 Hz), T92 (6.6 Hz), V95 (6.7 Hz), K96 (6.5 Hz), D98 (6.4 Hz), L100 (7.1 Hz), and K102 (8.9 Hz), which further supports the presence of helical conformations throughout sequence, aside from fraying at the termini.

Paramagnetic relaxation enhancement confirms absence of the broken‐helix conformation of aS in 30% HFIP

To investigate potential long‐range proximity of the N‐ and C‐termini of the lipid‐binding domain of aS in HFIP, as is observed in the presence of SDS micelles,9, 15 we examined the paramagnetic relaxation enhancement (PRE) of amide protons by nitroxide spin labels introduced into the protein at specific sites. The dipolar interactions between an unpaired electron in a nitroxide spin label and protons within ca. 20 Å of the label results in line broadening. Broadening can be assessed by comparing resonance intensities or linewidths for samples labeled with a paramagnetic spin label with those from diamagnetic samples in which the spin label is removed or reduced. Figure 6 shows the intensity ratios of HSQC cross‐peaks for paramagnetic and diamagnetic aS103stop samples in 30% HFIP compared with those observed previously in the presence of SDS micelles.15 In SDS micelle‐bound aS103stop, a spin label positioned at residue 83 or 93 causes broadening of resonances originating near the very N‐terminus of the protein and a spin label positioned at residue 9 leads to broadening near the C‐terminus of the membrane‐binding domain, indicating the proximity of the beginning of the first helix to the end of the second helix. In contrast, in 30% HFIP no corresponding broadening of aS103 resonances originating from regions distant to the spin label site is observed for spin labels at the same three locations.

Figure 6.

Figure 6

Open in a new tab

Paramagnetic Relaxation Enhancement for aS103stop in the presence of 30% HFIP (bar) and 40 mM SDS (red line) as previously reported.15 The data points represent the intensity ratios for each resolved cross‐peak in the 1H–15N HSQC spectra of nitroxide spin‐labeled and reduced protein samples.

In addition to spin labels placed near the termini of aS103, we also obtained data for a spin label attached at position 42, which is located right in the middle of the region that forms the non‐helical linker between helices 1 and 2 in the broken helix state. The broadening profile observed for this labeling site in HFIP is narrower than that observed in the micelle‐bound state, especially on the N‐terminal side of the labeling site. This suggests that the structure of the protein in this region is less compact and more extended in HFIP than in the presence of micelles. Together, the PRE profiles further demonstrate that aS does not adopt broken‐helix conformation in 30% HFIP, and instead adopts a more highly extended conformation, consistent with indications for a single extended helix conformation from the chemical shift and NOE data.

Dynamics suggest fast motions and helix fraying at the termini and slow flexing motions in the middle of the aS extended helix in HFIP

To determine whether the backbone dynamics of aS103stop in the presence of 30% HFIP shed any light on the conformations adopted by the protein, we performed measurements of backbone 15N R1 and R2 relaxation rates and the steady state heteronuclear 1H‐15N NOE (hNOE), which are sensitive to internal mobility and can provide information on the rapid (ps‐ns timescale) motions of individual residues, as well as on the presence of slower (us‐ms timescale) motions in some cases.46 Figure 7 shows these relaxation parameters as a function of residue number compared with values measured from SDS micelle bound aS.15 Both the R1 and R2 relaxation rates of micelle‐bound aS are relatively uniform away from the protein termini, showing a flattened pattern consistent with a single overall tumbling rate for the micelle‐bound protein. In contrast the R1 relaxation rates from aS103stop in 30% HFIP decrease gradually from residue 7 up to around residue 55 followed by a gradual increase till residue 90 resulting in a slightly concave curve with sharp drop of the R1 relaxation rate 3–4 residues from the termini. The R2 relaxation rates in 30% HFIP show a pattern opposite to that observed for R1, rising gradually up to around residue 55 followed by gradual decrease, again with sharp fall offs 3–4 residues from the termini.

Figure 7.

Figure 7

Open in a new tab

R1, R2, steady state 1H‐15N NOE relaxation parameters and R2/R1 for backbone 15N nuclei in aS103stop in the presence of 30% HFIP (bar) and 40 mM SDS (red line). R1, R2, steady state 1H‐15N NOE relaxation parameters for the SDS‐bound states are as previously reported.15

All the 1H‐15N heteronuclear NOE (hNOE) values of aS103stop in 30% HIFP are below 0.6, which is smaller than values of 0.8 or greater typically observed for well ordered proteins,47 indicating the presence of large amplitude NH bond vector motions on a sub‐nanosecond timescale throughout the protein. The hNOE values show a gradual increase from the N‐terminus up to residue 14, followed by a long plateau extending to residue 93 and ending with a dramatic decrease over the C‐terminal 6 residues. In addition to the expected flexibility at the termini, regions showing slightly lower hNOE values include residues 31–40 and 66–68. These correlate somewhat with regions exhibiting some lower secondary chemical shifts (residues 31–36 and 65–68) and weaker sequential amide proton NOE intensities (residues 31–34 and 64–67), indicating small increases in mobility may accompany reduced helical propensity in these regions.

The overall tumbling motion of a structured protein in solution is the major source of backbone NH bond reorientation and therefore of relaxation. It has been shown that the overall isotropic diffusion tensor can be determined from the R2/R1 ratio because the R2/R1 ratio is independent of internal motion and is highly sensitive to overall diffusion.48, 49, 50 In 30% HFIP, the R2/R1 ratios of aS103stop exhibit a maximum near the center of the protein with gradually decreasing values moving toward both termini [Fig. 7(D)]. Variation in the R2/R1 ratio can be caused either by anisotropic tumbling, under which the spin‐relaxation properties of a given 15N nucleus depend on the orientation of the NH bond within the protein (Tillett et al., 2007; Tjandra et al., 1995) or by the presence of slower motions associated with conformational or chemical exchange, which can make an additional contribution, Rex, to the transverse relaxation rate R2. Given the highly anisotropic nature of the extended helix structure implied by the chemical shift and NOE data for aSyn103stop in HFIP, the 15N relaxation data were analyzed using a model free analysis51 incorporating an anisotropic diffusion tensor, as implemented by the program TENSOR2.52 An initial optimized isotropic correlation time of 9.16 ns, calculated from the R2/R1 ratios after excluding residues with NOE values lower than 0.5 was used, together with initial estimates for a fully anisotropic diffusion tensor derived from a fully helical model for aS residues 1–94.6 The resulting generalized order parameter S2, a measure of NH bond vector motion amplitudes on the picoseconds to nanoseconds timescale, the exchange broadening factors Rex, arising from conformational exchange contributions occurring on a slower timescale of micro‐ to milliseconds, and τe, the effective internal correlation time for fast of internal motions, are plotted as a function of residue number in Figure 8. The S2 values are around 0.8 for residues 16–85, and fall off toward either terminus, suggesting a well‐ordered helical conformation at the center of the sequence with substantial fraying at the termini. Residues extending from the N‐ and C‐ termini to positions 35 and 65, respectively, require an internal correlation time to fit the data, suggesting they experience some degree of fast internal motion, which is greatest at the termini and falls off moving toward the center of the protein. Residues 5 to 85 also required an exchange factor, Rex for a proper fit of the data, indicating some degree of chemical exchange processes on slower timescales. The Rex terms are greatest at the center of the protein, with values around 20/S, and fall off in either direction. It is likely that these terms reflects structural plasticity rather than exchange between discrete conformations, as previously described.50 Together, these observations suggest that the long aS helix remains largely intact between residues 15 and 85, but may experience increasingly high frequency motions, suggestive of helix fraying, approaching the N and C‐termini. Slower time scale motions are also present, and are maximal at the center of the helix. These may reflect slow flexing motions of the helix about its center, which might be expected for such an elongated structure.

Figure 8.

Figure 8

Open in a new tab

Backbone dynamics parameters of aS103stop at 600 MHz obtained from the TENSOR2 analysis. Order parameters S2 (A), exchange contributions (Rex) to the 15N relaxation data (B) and correlation times of internal motions (τe) (C) as are plotted as a function of residue number.

Discussion

The interactions of aS with membranes, especially in the form of synaptic vesicles, are thought to be important for at least some of the normal functions of the protein and may also be important for aS aggregation and the pathogenesis of PD.26, 27, 53 Although synthetic phospholipid vesicles can closely mimic the size, curvature, and to some extent the composition of synaptic vesicles, and may therefore represent the best partner for investigating the structure of membrane‐bound aS, the application of high‐resolution solution NMR to vesicle‐bound synuclein is limited by the long correlation times of vesicles. Thus, high‐resolution solution NMR studies of lipid‐bound aS have largely been confined to micelles, where it was found that the N‐terminal lipid‐binding domain of aS adopts a helical conformation with a single break at residues 38–44, corresponding to the boundary between the first and second coding exons.6, 7, 9 However, it was argued that the high degree of curvature at the surface of micelles might generate a non‐physiological environment, which induces the break in the helical structure. Subsequent work using ESR and FRET distance measurements, as well as ESR spin label scanning indeed demonstrated that on the surface of vesicles, the lipid‐binding domain of aS adopts a single extended helix.17, 18, 19, 20 Nevertheless, other reports provided evidence that the broken‐helix state could be populated even on lower curvature surfaces,54, 55, 56 suggesting that the ability of aS to convert from the extended‐helix to the broken‐helix state may be ‘built in’ to the protein. A current limitation on our ability to assess the detailed structural properties of the extended‐helix state has been our inability to obtain high‐resolution data for this conformation of the protein. Here we use high concentrations of the FA HFIP to populate a state of C‐terminally truncated aS that is highly helical and resistant to aggregation,35, 36, 37, 38 properties that are shared by the membrane‐bound extended‐helix state, yet is amenable to the application of high‐resolution solution‐state NMR.

Information on secondary structure for aS in 30% HFIP was derived from carbon chemical shift, coupling constant and NOE measurements. These parameters indicate that aside from at the very termini, helical structure is continuous throughout the lipid‐binding domain of the protein in these conditions, confirming the formation of the extended‐helix state. Chemical shifts in particular indicate an increased helicity in the in the region that forms the non‐helical linker in the micelle‐bound state, namely residues 38–44. This increase in helicity is also consistent with the slightly increased helicity indicated in the CD data in HFIP compared to SDS micelles. Coupling constants also indicate a continuous helical structure throughout the protein, except for some fraying at the termini. Long‐range structure was also investigated using PRE data, which indicated that unlike the micelle‐bound broken‐helix state, the N‐ and C‐terminal ends of the lipid‐binding domain of aS do not come into detectable proximity in 30% HFIP. This is consistent with previous observations on the vesicle‐bound extended‐helix state, for which such distances have been shown to be quite long and to correspond to the distance expected for a continuous and highly extended helical structure.17, 18 The dynamic properties of aS103stop in 30% HFIP also support a highly elongated structure, as the relaxation data are clearly inconsistent with an isotropic tumbling mode. Instead, the data can be modeled using a highly anisotropic rotational diffusion tensor. The resulting fits are consistent with a single long helix that undergoes some of degree of fraying at its termini, and experiences slower flexing motions near its center. Similar behavior has been reported previously for apolipoproteins,50 which share sequence and structural features with aS.6

Despite strong evidence that aS103stop in 30% HFIP populates an extended helix conformation, it is notable that relatively weak sequential amide proton NOEs are seen for residues 41–45 and that NH‐NH(i,i + 2) NOEs for positions 42–44, 44–46, 45–47, and 46–48 and an Hα‐NH(i,i + 3) NOE for positions 42–45 are absent. These weak and absent short and medium range NOEs suggest the possibility of some intrinsic instability in the region that forms the linker in the broken helix state. This may be linked to the presence of residue Tyr39, which as the only aromatic residue present between positions 5 and 93 in the lipid‐binding domain of aS stands out as the most unique feature in the protein's primary sequence in this region.6 Indeed, recent studies of aS‐membrane interactions using nanodiscs show a very sharp boundary precisely at this position, with more stable membrane‐bound helical structure in the N‐terminal direction and significantly less stable structure in the C‐terminal direction.30 Such a sharp boundary is not observed in studies using vesicles,21, 29, 57 but this may be due to the much broader distribution of vesicle size, and hence membrane curvature, that is present in such assays. Since aS membrane binding is curvature sensitive, different binding curves corresponding to different vesicle sizes will be convolved together in such assays, potentially smearing out any sharp transitions. It is also notable that phosphorylation of Tyr39 by c‐Abl kinase, thought to be a functional modification of aS,58 enhances this transition from helical to non‐helical structure.59 Interestingly, PD‐associated mutations may also alter the location of this transition,28 and furthermore may influence the ability of aS to transition between the extended‐ and broken‐helix states.60 It will be interesting to examine the effects of PD mutations on the HFIP‐stabilized extended‐helix conformation of the protein.

In summary, high concentrations of FAs, and 30% HFIP in particular, are able to drive aS into a highly helical state extending from its N‐terminus to the end of its lipid‐binding domain. This state is distinct from the broken‐helix micelle‐bound conformation and instead shares properties with the vesicle‐bound extended helix state. Despite this difference, direct NMR measurements of this state suggest an intrinsic instability of the helical structure in the region that converts to a non‐helical linker upon micelle binding. This result supports other observations indicating that the lability of this region of the protein is functionally important, and that structural states in which the region corresponding to helix‐1 of the broken‐helix state remain bound to vesicles, while regions C‐terminal to this release from the vesicle membrane, possibly to bind to other membranes, may be both physiologically and pathologically relevant.27, 28, 30, 53, 61

Materials and Methods

Isotopically labeled proteins for NMR studies were produced from saturated overnight cultures used to inoculate M9 minimal media made with uniformly labeled 13C‐glucose and/or 15N‐ammonium chloride. Cultures were grown at 37°C to an A 600 nm of 0.5–0.6 at which point protein expression was induced with 1 mM IPTG. Cells were harvested by centrifugation 4 h post‐induction. Proteins were purified using a previously reported protocol involving ion‐exchange and reverse‐phase chromatography.11 Cell lysates were ultracentrifuged at 50,000 rpm in a Beckman Ti 50.2 rotor and the pellet was discarded. For further purification of the supernatant, a streptomycin sulfate cut and two ammonium sulfate cuts were used and the final pellet was resuspended in lysis buffer, dialyzed against 25 mM Tris, 20 mM NaCl, 3 M urea, 1 mM EDTA, applied to a cation exchange CE and eluted with a NaCl gradient. Protein containing fractions were pooled and dialyzed against 5% (v/v) acetic acid in water. Dialyzed sample was purified on a C4 reverse phase HPLC column using an acetonitrile and trifluoroacetic acid solvent system and lyophilized for storage at −20°C. Proteins eluted as a single peak and were highly pure as judged by the appearance of a single band in SDS/PAGE gels, a single peak in analytical HPLC chromatograms, and a single spectrum in subsequent NMR experiments. Protein mass and isotope labeling were confirmed by mass spectrometry.

Circular dichroism (CD) spectra were measured on an AVIV 62 DS spectrometer equipped with a sample temperature controller. Far‐UV CD spectra were monitored from 190 to 260 nm using final protein concentrations of 1 mg/mL with a path length of 0.2 mm, response time of 1 s, and scan speed of 50 nm/min. Protein concentrations were measured by UV absorbance using an estimated extinction coefficient of 1280 M−1 cm−1 at 280 nm. Protein samples were prepared under buffer conditions identical with those used in NMR experiments. Spectra were collected at 50°C.

NMR measurements were performed on samples dissolved in 100 mM NaCl, 10 mM Na2HPO4, pH 7.4 in 90%/10% H2O/D2O (NMR buffer) as previously described for full‐length or truncated aS6, 21 and filtered to remove any particulates or undissolved materials using microcon centrifugal filter devices (Millipore). HFIP was added into NMR buffer to achieve final concentrations of 30% v/v. NMR spectra were acquired on a 600 MHz Varian Unity INOVA spectrometer at 50°C. Pairs of HNCACB/CBCACONH and HNCACO/HNCO triple resonance experiments were collected to enable sequence‐specific backbone and Cβ resonance assignments for truncated aS in 30% HFIP. Secondary chemical shifts were calculated using the random coil shifts determined using hexapeptides in 1 M urea at pH ∼5.0.39 We have found that the use of these random coil shifts yields the most self‐consistent results for measurements under our conditions, as judged by fewer sporadic large secondary shifts, and a greater degree of contiguity within the data.62 NOE measurements included NOESY‐HSQC and HSQC‐NOESY‐HSQC experiments collected using 100 and 300 ms mixing times, respectively. 15N‐Edited HNHA spectra were collected to obtain 3JHNHA vicinal coupling constant values, with coupling constants calculated using the ratio of Hα crosspeak intensities to diagonal resonance intensities.63 A dephasing/rephrasing delay of 25 ms was used and no correction for selective Hα longitudinal relaxation rates was applied. Longitudinal (R1) and transverse (R2) relaxation rates for the backbone 15N nuclei were recorded using relaxation times of 10, 20, 40, 80, 160, 320, 640, 1280, and 1800 ms for R1 and 14.4, 28.8, 43.2, 57.6, 115.2, 230.4, and 460.8 ms for R2. To estimate noise levels, duplicate spectra were recorded for t = 80, and 640 ms (R1) and t = 14.4, and 115.2 ms (R2). R2 data were measured using a pulse sequence employing a Carr Purcell Meiboom Gill pulse train with a 900 μs delay between π pulses. R1 and R2 relaxation rates were determined by fitting resonance heights as a function of the relaxation delay time using NMRview.64 15N‐(1H) steady‐state heteronuclear NOE (HNOE) data were obtained as the ratio of peak heights in paired spectra collected with and without proton saturation during the relaxation delay of 5 s. The experimental uncertainty was estimated using the standard deviation of four pairs of spectra. All NMR data were processed with NMRPipe65 and analyzed using NMRView.64 Modelfree analysis was performed with the program TENSOR252 assuming axial symmetric rotational diffusion and a helical model of aS (residue 1–94).6

Site‐directed mutagenesis was used to substitute selected residues with cysteine. The purified mutant proteins were labeled with 1‐oxyl‐2,2,5,5‐tetramethylpyrroline‐3‐methyl‐methanehiosulfonate (MTSL, Toronto Research Chemicals) as described15 by adding a 10‐fold excess of spin label to freshly prepared samples of protein. For paramagnetic control samples, 1 mM DTT was added to the MTSL spin‐labeled sample to reduce the nitroxide spin label from the protein. Paramagnetic relaxation enhancement was measured by collecting 1H‐15N HSQC spectra using 0.25 mM protein samples at 50°C. The intensities of cross peaks in the 1H‐15N HSQC spectra of both the spin labeled and reduced samples were measured and their ratio calculated and plotted.

Conflict of Interest

The authors have no conflicts to declare.

References

  • 1. Jensen PH, Nielsen MS, Jakes R, Dotti CG, Goedert M (1998) Binding of alpha‐synuclein to brain vesicles is abolished by familial Parkinson's disease mutation. J Biol Chem 273:26292–26294. [DOI] [PubMed] [Google Scholar]
  • 2. Fortin DL, Troyer MD, Nakamura K, Kubo S, Anthony MD, Edwards RH (2004) Lipid rafts mediate the synaptic localization of alpha‐synuclein. J Neurosci 24:6715–6723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Cole NB, Murphy DD, Grider T, Rueter S, Brasaemle D, Nussbaum RL (2002) Lipid droplet binding and oligomerization properties of the Parkinson's disease protein alpha‐synuclein. J Biol Chem 277:6344–6352. [DOI] [PubMed] [Google Scholar]
  • 4. Outeiro TF, Lindquist S (2003) Yeast cells provide insight into alpha‐synuclein biology and pathobiology. Science 302:1772–1775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Davidson WS, Jonas A, Clayton DF, George JM (1998) Stabilization of alpha‐synuclein secondary structure upon binding to synthetic membranes. J Biol Chem 273:9443–9449. [DOI] [PubMed] [Google Scholar]
  • 6. Bussell R, Jr , Eliezer D (2003) A structural and functional role for 11‐mer repeats in alpha‐synuclein and other exchangeable lipid binding proteins. J Mol Biol 329:763–778. [DOI] [PubMed] [Google Scholar]
  • 7. Chandra S, Chen X, Rizo J, Jahn R, Sudhof TC (2003) A broken alpha‐helix in folded alpha‐Synuclein. J Biol Chem 278:15313–15318. [DOI] [PubMed] [Google Scholar]
  • 8. Jao CC, Der‐Sarkissian A, Chen J, Langen R (2004) Structure of membrane‐bound alpha‐synuclein studied by site‐directed spin labeling. Proc Natl Acad Sci USA 101:8331–8336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Ulmer TS, Bax A, Cole NB, Nussbaum RL (2005) Structure and dynamics of micelle‐bound human alpha‐synuclein. J Biol Chem 280:9595–9603. [DOI] [PubMed] [Google Scholar]
  • 10. Rhoades E, Ramlall TF, Webb WW, Eliezer D (2006) Quantification of alpha‐synuclein binding to lipid vesicles using fluorescence correlation spectroscopy. Biophys J 90:4692–4700. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Eliezer D, Kutluay E, Bussell R, Jr , Browne G (2001) Conformational properties of alpha‐synuclein in its free and lipid‐ associated states. J Mol Biol 307:1061–1073. [DOI] [PubMed] [Google Scholar]
  • 12. Bussell R, Jr , Eliezer D (2001) Residual structure and dynamics in Parkinson's disease‐associated mutants of alpha‐synuclein. J Biol Chem 276:45996–46003. [DOI] [PubMed] [Google Scholar]
  • 13. Uversky VN, Li J, Souillac P, Millett IS, Doniach S, Jakes R, Goedert M, Fink AL (2002) Biophysical properties of the synucleins and their propensities to fibrillate: inhibition of alpha‐synuclein assembly by beta‐ and gamma‐synucleins. J Biol Chem 277:11970–11978. [DOI] [PubMed] [Google Scholar]
  • 14. Weinreb PH, Zhen W, Poon AW, Conway KA, Lansbury PT Jr (1996) NACP, a protein implicated in Alzheimer's disease and learning, is natively unfolded. Biochemistry 35:13709–13715. [DOI] [PubMed] [Google Scholar]
  • 15. Bussell R, Jr , Ramlall TF, Eliezer D (2005) Helix periodicity, topology, and dynamics of membrane‐associated alpha‐synuclein. Protein Sci 14:862–872. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Borbat P, Ramlall TF, Freed JH, Eliezer D (2006) Inter‐helix distances in lysophospholipid micelle‐bound alpha‐synuclein from pulsed ESR measurements. J Am Chem Soc 128:10004–10005. [DOI] [PubMed] [Google Scholar]
  • 17. Georgieva ER, Ramlall TF, Borbat PP, Freed JH, Eliezer D (2008) Membrane‐bound alpha‐synuclein forms an extended helix: long‐distance pulsed ESR measurements using vesicles, bicelles, and rodlike micelles. J Am Chem Soc 130:12856–12857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Trexler A, Rhoades E (2009) Alpha‐Synuclein binds large unilamellar vesicles as an extended helix. Biochemistry 48:2304–2306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Jao CC, Hegde BG, Chen J, Haworth IS, Langen R (2008) Structure of membrane‐bound alpha‐synuclein from site‐directed spin labeling and computational refinement. Proc Natl Acad Sci USA 105:19666–19671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Ferreon AC, Gambin Y, Lemke EA, Deniz AA (2009) Interplay of alpha‐synuclein binding and conformational switching probed by single‐molecule fluorescence. Proc Natl Acad Sci USA 106:5645–5650. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Bussell R, Jr , Eliezer D (2004) Effects of Parkinson's disease‐linked mutations on the structure of lipid‐associated alpha‐synuclein. Biochemistry 43:4810–4818. [DOI] [PubMed] [Google Scholar]
  • 22. Kim YS, Laurine E, Woods W, Lee SJ (2006) A novel mechanism of interaction between alpha‐synuclein and biological membranes. J Mol Biol 360:386–397. [DOI] [PubMed] [Google Scholar]
  • 23. Fortin DL, Nemani VM, Voglmaier SM, Anthony MD, Ryan TA, Edwards RH (2005) Neural activity controls the synaptic accumulation of alpha‐synuclein. J Neurosci 25:10913–10921. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Ulmer TS, Bax A (2005) Comparison of structure and dynamics of micelle‐bound human alpha‐synuclein and Parkinson disease variants. J Biol Chem 280:43179–43187. [DOI] [PubMed] [Google Scholar]
  • 25. Burre J, Sharma M, Sudhof TC (2014) alpha‐Synuclein assembles into higher‐order multimers upon membrane binding to promote SNARE complex formation. Proc Natl Acad Sci USA 111:E4274–E4283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Eliezer D Protein folding and aggregation in in vitro models of Parkinson's disease: structure and function of α–Synuclein In: Nass R. & Prezedborski S, Eds. (2008) Parkinson's disease: molecular and therapeutic insights from model systems. New York: Academic Press, pp 575–595. [Google Scholar]
  • 27. Dikiy I, Eliezer D (2012) Folding and misfolding of alpha‐synuclein on membranes. Biochim Biophys Acta 1818:1013–1018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Fusco G, Pape T, Stephens AD, Mahou P, Costa AR, Kaminski CF, Schierle GSK, Vendruscolo M, Veglia G, Dobson CM, De Simone A (2016) Structural basis of synaptic vesicle assembly promoted by alpha‐synuclein. Nat Commun 7:12563. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Bodner CR, Dobson CM, Bax A (2009) Multiple tight phospholipid‐binding modes of alpha‐Synuclein revealed by solution NMR spectroscopy. J Mol Biol 390:775–790. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Viennet T, Wördehoff MM, Uluca B, Poojari C, Shaykhalishahi H, Willbold D, Strodel B, Heise H, Buell AK, Hoyer W, Etzkorn M (2017) A structural and kinetic link between membrane association and amyloid fibril formation of α‐Synuclein. bioRxiv. https://doi.org/10.1101/173724 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Fioroni M, Diaz MD, Burger K, Berger S (2002) Solvation phenomena of a tetrapeptide in water/trifluoroethanol and water/ethanol mixtures: a diffusion NMR, intermolecular NOE, and molecular dynamics study. J Am Chem Soc 124:7737–7744. [DOI] [PubMed] [Google Scholar]
  • 32. Gerig JT (2004) Structure and solvation of melittin in 1,1,1,3,3,3‐hexafluoro‐2‐propanol/water. Biophys J 86:3166–3175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Roccatano D, Colombo G, Fioroni M, Mark AE (2002) Mechanism by which 2,2,2‐trifluoroethanol/water mixtures stabilize secondary‐structure formation in peptides: a molecular dynamics study. Proc Natl Acad Sci USA 99:12179–12184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Eliezer D, Barre P, Kobaslija M, Chan D, Li X, Heend L (2005) Residual structure in the repeat domain of tau: echoes of microtubule binding and paired helical filament formation. Biochemistry 44:1026–1036. [DOI] [PubMed] [Google Scholar]
  • 35. Li HT, Du HN, Tang L, Hu J, Hu HY (2002) Structural transformation and aggregation of human alpha‐synuclein in trifluoroethanol: non‐amyloid component sequence is essential and beta‐sheet formation is prerequisite to aggregation. Biopolymers 64:221–226. [DOI] [PubMed] [Google Scholar]
  • 36. Munishkina LA, Phelan C, Uversky VN, Fink AL (2003) Conformational behavior and aggregation of alpha‐synuclein in organic solvents: modeling the effects of membranes. Biochemistry 42:2720–2730. [DOI] [PubMed] [Google Scholar]
  • 37. Kuroda Y, Maeda Y, Hanaoka H, Miyamoto K, Nakagawa T (2004) Oligopeptide‐mediated acceleration of amyloid fibril formation of amyloid beta(Abeta) and alpha‐synuclein fragment peptide (NAC). J Pept Sci 10:8–17. [DOI] [PubMed] [Google Scholar]
  • 38. Anderson VL, Ramlall TF, Rospigliosi CC, Webb WW, Eliezer D (2010) Identification of a helical intermediate in trifluoroethanol‐induced alpha‐synuclein aggregation. Proc Natl Acad Sci USA 107:18850–18855. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Wishart DS, Bigam CG, Holm A, Hodges RS, Sykes BD (1995) 1H, 13C, and 15N random coil NMR chemical shifts of the common amino acids. I. Investigations of nearest‐neighbor effects. J Biomol NMR 5:67–81. [DOI] [PubMed] [Google Scholar]
  • 40. Wishart DS, Sykes BD (1994) The 13C chemical‐shift index: a simple method for the identification of protein secondary structure using 13C chemical‐shift data. J Biomol NMR 4:171–180. [DOI] [PubMed] [Google Scholar]
  • 41. Wishart DS, Case DA (2002) Use of chemical shifts in macromolecular structure determination. Methods Enzymol 338:3–34. [DOI] [PubMed] [Google Scholar]
  • 42. Pardi A, Billeter M, Wuthrich K (1984) Calibration of the angular dependence of the amide proton‐C alpha proton coupling constants, 3JHN alpha, in a globular protein. Use of 3JHN alpha for identification of helical secondary structure. J Mol Biol 180:741–751. [DOI] [PubMed] [Google Scholar]
  • 43. Wuthrich K (1986) NMR of proteins and nucleic acids. New York: J. Wiley & Sons, p 292. [Google Scholar]
  • 44. Smith LJ, Bolin KA, Schwalbe H, MacArthur MW, Thornton JM, Dobson CM (1996) Analysis of main chain torsion angles in proteins: prediction of NMR coupling constants for native and random coil conformations. J Mol Biol 255:494–506. [DOI] [PubMed] [Google Scholar]
  • 45. Kessler H, Griesinger C, Lautz J, Mueller A, Van Gunsteren WF, Berendsen HJC (1988) Conformational dynamics detected by nuclear magnetic resonance NOE values and J coupling constants. J Am Chem Soc 110:3393–3396. [Google Scholar]
  • 46. Palmer AG, 3rd , Kroenke CD, Loria JP (2001) Nuclear magnetic resonance methods for quantifying microsecond‐to‐millisecond motions in biological macromolecules. Methods Enzymol 339:204–238. [DOI] [PubMed] [Google Scholar]
  • 47. Grasberger BL, Gronenborn AM, Clore GM (1993) Analysis of the backbone dynamics of interleukin‐8 by 15N relaxation measurements. J Mol Biol 230:364–372. [DOI] [PubMed] [Google Scholar]
  • 48. Kay LE, Torchia DA, Bax A (1989) Backbone dynamics of proteins as studied by 15N inverse detected heteronuclear NMR spectroscopy: application to staphylococcal nuclease. Biochemistry 28:8972–8979. [DOI] [PubMed] [Google Scholar]
  • 49. Brutscher B, Bruschweiler R, Ernst RR (1997) Backbone dynamics and structural characterization of the partially folded A state of ubiquitin by 1H, 13C, and 15N nuclear magnetic resonance spectroscopy. Biochemistry 36:13043–13053. [DOI] [PubMed] [Google Scholar]
  • 50. Raussens V, Slupsky CM, Ryan RO, Sykes BD (2002) NMR structure and dynamics of a receptor‐active apolipoprotein E peptide. J Biol Chem 277:29172–29180. [DOI] [PubMed] [Google Scholar]
  • 51. Lipari G, Szabo A (1982) Model‐free approach to the interpretation of nuclear magnetic resonance relaxation in macromolecules. 2. Analysis of experimental results. J Am Chem Soc 104:4559–4570. [Google Scholar]
  • 52. Dosset P, Hus JC, Blackledge M, Marion D (2000) Efficient analysis of macromolecular rotational diffusion from heteronuclear relaxation data. J Biomol NMR 16:23–28. [DOI] [PubMed] [Google Scholar]
  • 53. Snead D, Eliezer D (2014) Alpha‐synuclein function and dysfunction on cellular membranes. Exp Neurobiol 23:292–313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Drescher M, Veldhuis G, van Rooijen BD, Milikisyants S, Subramaniam V, Huber M (2008) Antiparallel arrangement of the helices of vesicle‐bound alpha‐synuclein. J Am Chem Soc 130:7796–7797. [DOI] [PubMed] [Google Scholar]
  • 55. Bortolus M, Tombolato F, Tessari I, Bisaglia M, Mammi S, Bubacco L, Ferrarini A, Maniero AL (2008) Broken helix in vesicle and micelle‐bound alpha‐synuclein: insights from site‐directed spin labeling‐EPR experiments and MD simulations. J Am Chem Soc 130:6690–6691. [DOI] [PubMed] [Google Scholar]
  • 56. Robotta M, Braun P, van Rooijen B, Subramaniam V, Huber M, Drescher M (2011) Direct evidence of coexisting horseshoe and extended helix conformations of membrane‐bound alpha‐synuclein. Chemphyschem 12:267–269. [DOI] [PubMed] [Google Scholar]
  • 57. Fusco G, De Simone A, Gopinath T, Vostrikov V, Vendruscolo M, Dobson CM, Veglia G (2014) Direct observation of the three regions in alpha‐synuclein that determine its membrane‐bound behaviour. Nat Commun 5:3827. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58. Mahul‐Mellier AL, Fauvet B, Gysbers A, Dikiy I, Oueslati A, Georgeon S, Lamontanara AJ, Bisquertt A, Eliezer D, Masliah E, Halliday G, Hantschel O, Lashuel HA (2014) c‐Abl phosphorylates alpha‐synuclein and regulates its degradation: implication for alpha‐synuclein clearance and contribution to the pathogenesis of Parkinson's disease. Hum Mol Genet 23:2858–2879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59. Dikiy I, Fauvet B, Jovicic A, Mahul‐Mellier AL, Desobry C, El‐Turk F, Gitler AD, Lashuel HA, Eliezer D (2016) Semisynthetic and in vitro phosphorylation of alpha‐Synuclein at Y39 promotes functional partly helical membrane‐bound states resembling those induced by PD mutations. ACS Chem Biol 11:2428–2437. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60. Georgieva ER, Ramlall TF, Borbat PP, Freed JH, Eliezer D (2010) The lipid‐binding domain of wild type and mutant alpha‐synuclein: compactness and interconversion between the broken and extended helix forms. J Biol Chem 285:28261–28274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61. Robotta M, Hintze C, Schildknecht S, Zijlstra N, Jungst C, Karreman C, Huber M, Leist M, Subramaniam V, Drescher M (2012) Locally resolved membrane binding affinity of the N‐terminus of alpha‐synuclein. Biochemistry 51:3960–3962. [DOI] [PubMed] [Google Scholar]
  • 62. Eliezer D (2007) Characterizing residual structure in disordered protein States using nuclear magnetic resonance. Methods Mol Biol 350:49–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63. Vuister GW, Bax A (1993) Quantitative J correlation: a new approach for measuring homonuclear three‐bond (JHNHa) coupling constants in 15N‐enriched proteins. J Am Chem Soc 115:7772–7777. [Google Scholar]
  • 64. Johnson BA, Blevins RA (1994) NMRView: a computer program for the visualization and analysis of NMR data. J Biomol NMR 4:603–614. [DOI] [PubMed] [Google Scholar]
  • 65. Delaglio F, Grzesiek S, Vuister GW, Zhu G, Pfeifer J, Bax A (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes [see comments]. J Biomol NMR 6:277–293. [DOI] [PubMed] [Google Scholar]
  • 66. Sung YH, Eliezer D (2007) Residual structure, backbone dynamics, and interactions within the Synuclein family. J Mol Biol 372:689–707. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of The Protein Society

RESOURCES