Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Sep 19.
Published in final edited form as: J Am Chem Soc. 2026 Sep 2;148(34):37102–37109. doi: 10.1021/jacs.6c10987

Proteoform-Resolved Cross-Linking Reveals Environment-Dependent Structural Effects of α-Synuclein S129 Phosphorylation

Ashlyn N Dollar 1, Daniel W Kelley 2, Ian K Webb 3,*
PMCID: PMC13587262  NIHMSID: NIHMS2207809  PMID: 42701266

Abstract

Intrinsically disordered proteins (IDPs) drive many neurodegenerative disorders, but their structures remain difficult to define because they populate dynamic ensembles that change with the chemical environment. This problem is central for α-synuclein (aSyn), a Parkinson’s disease-linked IDP in which S129 phosphorylation is highly enriched in disease-associated aggregates. Although S129 phosphorylation stabilizes a more compact aSyn ensemble in dilute solution, whether this structural effect persists across other biochemical environments has remained unknown. Here, we developed a quantitative cross-linking mass spectrometry framework to determine how WT and pS129 aSyn respond to chemically distinct environments. We compared dilute buffer with two perturbations relevant to aSyn biology: trimethylamine N-oxide (TMAO), a gut-microbiome-derived metabolite associated with Parkinson’s disease that can also act as a compacting osmolyte at high concentration, and octyl glucoside (OG) micelles, which provide a membrane-mimetic surface. TMAO rewired the phosphorylation-dependent structural response in a concentration-dependent manner: 1.8 M TMAO shifted WT aSyn toward the dilute pS129 contact pattern by increasing long-range contacts between the N-terminal and C-terminal regions, whereas 3.4 M TMAO collapsed both proteoforms and reduced their structural differences. In OG micelles, the proteoforms diverged. WT favored an extended-helix-like contact pattern with stronger contacts between the C-terminal tail and micelle-bound N-terminal region, whereas pS129 favored contacts between the N-terminal and NAC regions and fewer contacts to the C-terminal tail, consistent with a broken-helix-like topology. By integrating regional contact counts, normalized cross-link intensities, and geometric compatibility analysis, this workflow distinguishes broad contact accessibility from the residue-level contacts that dominate each condition. These results establish a proteoform-by-environment model for IDP structure, in which phosphorylation effects are not fixed but are rewritten by the surrounding chemical environment.

Graphical Abstract

graphic file with name nihms-2207809-f0001.webp

INTRODUCTION

Intrinsically disordered proteins (IDPs) do not adopt a single stable tertiary structure. Instead, they populate conformational ensembles that are sensitive to sequence, post-translational modification, binding partners, and solution environment.1-3 α-Synuclein (aSyn) is a small, acidic IDP involved in synaptic signaling that exists primarily as an unfolded monomer in mammalian cells,4,5 but it can also form aggregates associated with Parkinson’s disease (PD), Lewy body dementia, and multiple system atrophy (MSA).6 aSyn (Figure 1) contains an N-terminal amphipathic region that becomes α-helical upon lipid binding (NT, residues 1–60),7 a hydrophobic non-amyloid-β-component region (NAC, residues 61–95), and an acidic C-terminal region that remains highly disordered (CT, residues 96–140).8,9 Phosphorylation at serine 129 (pS129) is strongly associated with pathology, and more than 90% of aSyn in Lewy body lesions is phosphorylated at this site, compared with less than 5% in unaffected brain tissue.10-12 Because ensemble-averaging methods can obscure low-population and transient contacts, we used cross-linking mass spectrometry to compare long-range contacts in wild-type (WT) and pS129 aSyn across distinct chemical environments.

Figure 1.

Figure 1.

Open in a new tab

Structural regions of human aSyn, including the amphipathic N-terminal region, the NAC, and the acidic C-terminal region. Helical features from the broken helix micelle-bound structure are annotated.48

S129 phosphorylation has been linked to both disease and normal regulation of neurotransmission,13,14 but its effects on aSyn aggregation, toxicity, and structure remain debated.12,15-26 In dilute, buffered solutions, phosphorylation shifts aSyn toward more compact conformational ensembles.27-32 However, dilute solution does not capture key chemical features that shape aSyn in cells, including excluded-volume effects,33 osmolytes,34 and membrane surfaces.35 Trimethylamine N-oxide (TMAO) is a small osmolyte that is excluded from protein surfaces and can favor compact or folded conformations.36-39 TMAO is also relevant to aSyn biology because it is a gut-microbiome-derived metabolite detected in plasma and brain, and altered TMAO levels have been associated with PD status, severity, or progression in metabolomic and clinical studies.40-43 At 3.4 M, TMAO has also been shown to induce rapid folding of WT-aSyn.44,45 In contrast, lipid membranes and detergent micelles promote α-helical structure in the N-terminal region, with aSyn adopting either an extended helix or a broken two-helix topology depending on the membrane mimic and binding conditions.46-49

Cross-linking mass spectrometry (XL-MS) has previously been used with DOPS liposomes to probe membrane-bound aSyn, where most detected cross-links were within the N-terminal region and only one link connected the N- and C-terminal regions.50 We previously used cross-linking and covalent labeling MS to show that pS129-aSyn is more compact than WT-aSyn in dilute solution, with a more compact N-terminal region and more long-range contacts.32 However, the conformational response of pS129-aSyn to crowded or membrane-mimetic environments has not been measured.

Here, we developed a proteoform-resolved quantitative XL-MS workflow to test whether S129 phosphorylation produces an intrinsic compaction of aSyn or instead changes how the ensemble responds to the chemical environment. WT and pS129 aSyn were analyzed in dilute buffer, 1.8 and 3.4 M TMAO, and 50 mM octyl glucoside (OG). These conditions span a conventional dilute reference state, osmolyte-driven compaction by TMAO, and membrane-mimetic binding to OG micelles. TMAO is especially relevant because it is both a compacting osmolyte and a gut-microbiome-derived metabolite linked to Parkinson’s disease through metabolomic and clinical studies. OG micelles provide a complementary environment that promotes N-terminal helix formation and mimics key features of membrane binding. Rather than treating cross-links only as present-or-absent distance restraints, we quantified cross-link intensities and compared regional contact counts with residue-level contact patterns across conditions. This design allowed us to ask not only which contacts can form but which contacts dominate each proteoform in each environment. The results support a proteoform-by-environment model for aSyn structure: S129 phosphorylation does not simply make aSyn compact but rewires how the ensemble responds to solution compaction and membrane-mimetic binding. More broadly, this work shows that PTM effects in IDPs must be interpreted in the chemical environments where those proteoforms function.

RESULTS AND DISCUSSION

To compare environment-dependent changes in aSyn structure, we used three cross-linking chemistries that report different distance and residue constraints. Zero-length cross-linking captures close amine–carboxylate contacts, often reflecting electrostatic proximity.51-54 BS2G and BS3 are amine-reactive cross-linkers with spacer arms of 7.7 Å and 11.4 Å, respectively, and probe contacts involving the N-terminus and lysine residues through K102. Because zero-length cross-linking also targets acidic residues, it extends coverage into the disordered C-terminal region that contains S129. Cross-links identified in at least two of three replicates are shown in Figure 2. These residue-level cross-link maps provide the experimental basis for the major conformational families summarized in Figure 4.

Figure 2.

Figure 2.

Open in a new tab

Cross-linked residue pairs identified for WT and pS129 aSyn in dilute solution, 1.8 M TMAO, 3.4 M TMAO, and 50 mM OG using zero-length, BS2G, and BS3 cross-linking. Only links identified in at least two of three replicates are shown. We previously reported the dilute-solution cross-linking.32

Figure 4.

Figure 4.

Open in a new tab

Contact models of dominant conformational families supported by the XL-MS data. Colors are chosen to match Figure 1 (dark blue is NT, black is NAC, and red is CT). The regional cross-links are summarized by connecting lines. Line colors compare each cross-linker class across conditions, not across different contact classes. Red lines indicate a high relative percentage for a given regional contact, and black lines indicate low percentages. Stars represent phosphorylation of S129. A. WT in dilute solution is most consistent with extended conformations. B. pS129 in dilute solution, WT in 1.8 M TMAO, and pS129 in 1.8 M TMAO show U-turn-like contact patterns. C. High TMAO supports collapsed contact patterns for both proteoforms. D. OG supports an extended-helix-like contact pattern for WT and E. a broken-helix-like contact pattern with a dynamic C-terminal tail for pS129.

1.8. M TMAO Partially Compacts WT but Has Little Additional Effect on pS129

TMAO is a naturally occurring osmolyte that is excluded from protein surfaces and can entropically favor compact conformations.39 It is also useful for XL-MS because it can be removed by molecular weight cutoff filtration after cross-linking. At 1.8 M TMAO, WT aSyn showed only modest changes relative to dilute solution (Figure 2). To compare regional contact patterns, each unique cross-link was assigned to the pair of protein regions it connected: NT–NT, NT–NAC, NT–CT, NAC–NAC, NAC–CT, or CT–CT (Figure 3). Each link was then treated as present or absent, and the number of links in each regional class was compared across conditions using chi-squared tests for independence (Data set S1). Cross-link intensities were analyzed separately in the PCA analysis described below. From dilute solution to 1.8 M TMAO, WT aSyn gained five NT–CT cross-links, including a zero-length link between the N-terminus and E139. This shift is consistent with partial compaction and with the U-turn-like conformer family summarized in Figure 4B, although the category-level change was only nominally significant (p = 0.026). In contrast, pS129 showed no significant category-level change relative to dilute solution (p > 0.05). Therefore, 1.8 M TMAO changed WT more than pS129, consistent with pS129 already sampling compact NT–CT contacts in dilute solution.

Figure 3.

Figure 3.

Open in a new tab

Regional cross-linking patterns across proteoform and environment. A. Fraction of cross-links assigned to each regional class. B-D. NT-NT, NT-NAC, and NT-CT cross-links as a function of proteoform and experimental condition. Significance was determined by chi-squared tests for independence. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001. Reported p values are uncorrected and are used as category-level comparisons. Statistics were performed on cross-link counts.

Figure 3 provides the regional cross-linking patterns that were used as the first level of analysis. Figure 3A shows the full distribution of detected cross-links across regional classes for each proteoform and environment. Figures 3B-D then isolate the three classes that changed most strongly across conditions: NT–NT, NT–NAC, and NT–CT contacts. Figure 3D shows that in dilute buffer, pS129 had significantly more NT–CT cross-links than WT, consistent with phosphorylation-driven compaction (Figure 4).32 In 1.8 M TMAO, however, the WT and pS129 regional profiles became more similar. Although NT–NT cross-links remained dominant for both proteoforms (Figure 3A,B), Figures 3C and D show that the numbers of NT–NAC and NT–CT cross-links were not significantly different between WT and pS129. These data suggest that 1.8 M TMAO shifts WT aSyn toward contact patterns already accessible to pS129 in dilute solution.

This modest compaction is consistent with PRE-NMR studies showing that WT aSyn remains highly disordered in mammalian cells but adopts a slightly smaller ensemble than it does in dilute buffer.33 A similar ~ 10% reduction in radius of gyration was reproduced in vitro with BSA and Ficoll crowding agents. Therefore, the 1.8 M TMAO data are consistent with the broader observation that nondilute environments can modestly compact WT aSyn. This comparison should not be interpreted as TMAO and macromolecular crowders acting by the same mechanism. Rather, it shows that phosphorylation-dependent structural differences observed in dilute buffer can be reduced when the solution environment favors compaction.

3.4. M TMAO Drives Extensive Compaction for Both Proteoforms

Increasing TMAO to 3.4 M changed the cross-linking patterns dramatically. Both proteoforms showed absolute increases across all regional categories and a higher percentage of NT–CT cross-links (Figure 3A,D), consistent with highly compact ensembles. NT–CT links were strongly and significantly enriched compared to 1.8 M TMAO. The number of cross-links increased from 5 at 1.8 M TMAO to 47 at 3.4 M TMAO for WT and from 5 to 43 for pS129. At the same time, the fractional contribution of NT–NT cross-links decreased (Figure 3B), even though the total number of detected cross-links nearly doubled. For WT, NT–NT links changed from 72/90 at 1.8 M TMAO to 75/161 at 3.4 M TMAO. For pS129, they changed from 53/72 to 65/141. No significant changes were observed for NT–NAC cross-links. The shifts in cross-linking patterns indicate that 3.4 M TMAO brings distant regions, especially the NT and CT regions, into close proximity, supporting the collapsed conformational family shown in Figure 4C. The collapse is qualitatively consistent with simulations55 and spectroscopic measurements showing that increasing TMAO concentration decreases peptide radius of gyration and slows conformational dynamics.37

At 3.4 M TMAO, WT and pS129 did not show significant differences in the regional cross-linking patterns. This suggests that strong osmolyte-driven collapse reduces the proteoform-specific differences that are apparent in a dilute solution. In this high-TMAO limit, the chemical environment dominates the detectable intramolecular contact patterns.

OG Micelles Separate WT and pS129 into Distinct Membrane-Mimetic Contact Patterns

OG micelles provide a curved amphipathic surface that promotes α-helical binding of aSyn and is commonly used as a membrane mimic.56 Binding conditions can favor either an extended helix or a broken two-helix topology.49 Compared to dilute solution, WT aSyn in OG showed a strong enrichment of NT–CT cross-links, increasing from 0 to 53 links, while the fraction of NT–NT cross-links decreased from 71/87 to 92/174 (Figure 3A). NT–NAC cross-links also decreased modestly from 5 to 2 (Figure 3C). These changes are consistent with the extended-helix-like micelle-bound state49 in Figure 4D, in which the C-terminal tail remains near the micelle-bound N-terminal region. Compared with 3.4 M TMAO, OG produced fewer NT–NAC cross-links (Figure 3C), suggesting that 3.4 M TMAO and OG both favor compact contact patterns but do so through different structural routes. These OG- and TMAO-associated states are also distinct from the elongated conformations observed by XL-MS under liquid–liquid phase separation conditions.54

XL-MS patterns for pS129 aSyn in OG were strikingly different from WT. Although the total number of cross-links was similar for both proteoforms (174 for WT and 184 for pS129), pS129 had 35 more NT–NT cross-links, 15 more NT–NAC cross-links, and 29 fewer NT–CT cross-links (Figure 3B-D). This shift places the NT and NAC regions in closer proximity while reducing contacts between the NT region and CT tail, consistent with the broken-helix-like micelle-bound topology in Figure 4E.49 Figure 2 shows increased cross-linking between K80 in the NAC region and the NT region for pS129, while no such links were detected for the WT. For example, the K12–K80 cross-link was unique to pS129 in OG and supports proximity between the early N-terminal region and the NAC region.57 The reduction in NT–CT cross-links suggests that the C-terminal tail is less collapsed against the micelle-bound helical region in pS129 than in WT. Compared with dilute solution and 1.8 M TMAO, pS129 in OG showed increased NT–NAC cross-linking. Compared with 3.4 M TMAO, pS129 in OG showed reduced NT–CT cross-linking (Figure 3D). Together, these differences support a micelle-bound pS129 ensemble with stronger NT–NAC contacts and a more dynamic C-terminal tail (Figure 4E).

Quantitative XL-MS Separates Accessible Contacts from Dominant Conformational States

To determine whether residue-level cross-link intensities supported the trends inferred from regional contact counts, we performed PCA with k-means clustering using normalized intensities of cross-links identified in at least two of the three replicates for both proteoforms across the four solution conditions (Figure 5). This analysis compared global cross-linking profiles rather than individual links. The first two principal components explained 37.7% of the variance and were used for visualization, while the first five accounted for 73.0% of the variance. Thus, the normalized XL-MS data contained structured variation that could be summarized with a small number of components. The resulting score plot separated several proteoform- and environment-dependent profiles, indicating that cross-link intensities provided information beyond the regional contact count analysis. WT dilute, WT 1.8 M TMAO, and pS129 1.8 M TMAO clustered together, whereas pS129 dilute separated along PC1. Although regional contact-count analysis showed that WT in 1.8 M TMAO and pS129 in dilute solution shared some broad contact classes, PCA showed that their most intense residue-level cross-links remained distinct. 1.8 M TMAO therefore made WT resemble pS129 at the level of broad regional contacts, but it does not reproduce the dilute pS129 ensemble.

Figure 5.

Figure 5.

Open in a new tab

PCA with k-means clustering based on normalized intensities of quantified cross-linked peptides across three replicates for each condition. The number of principal components was selected from an elbow plot, and the number of clusters was selected from silhouette analysis.

The two 3.4 M TMAO samples separated from the dilute and 1.8 M TMAO samples along PC2, consistent with the large increase in long-range contacts observed with 3.4 M TMAO. The OG samples occupied a different region of the PCA plot, indicating that micelle binding produces a distinct cross-linking pattern. WT and pS129 also separated from each other in OG, supporting different micelle-bound conformations (Figure 4). The regional contact-count analysis suggested that WT 3.4 M TMAO, pS129 3.4 M TMAO, and pS129 OG share some broad contact classes. However, this similarity does not mean that the same ensemble is populated in each condition. In 3.4 M TMAO, both proteoforms collapse in solution and form many NT–CT and NT–NAC contacts. In OG, pS129 reaches some of the same broad regional contact classes through micelle binding, but with fewer NT–CT contacts and stronger NT–NAC contacts. Hence, regional contact counts show which types of contacts are detected, whereas PCA of normalized intensities separates conditions based on which specific residue-level contacts dominate.

To connect the quantitative XL-MS data to possible conformational substates, we performed a geometric compatibility analysis using a library of 10,000 aSyn conformers generated with STARLING.58 STARLING is a recently reported deep-learning approach that generates conformational ensembles of intrinsically disordered proteins from sequence. Here, STARLING was used only to provide a diverse conformer library; the compatibility analysis was applied separately to determine which high-intensity XL-MS restraints could be satisfied by the same individual conformer. For each condition, the most intense normalized cross-links were compared with each conformer using the distance criteria described in the Supporting Information. This analysis was not intended to determine a single structure or imply that all detected links formed on the same molecule. Instead, compatible links were treated as residue-level contact sets that could plausibly belong to the same conformational substate. Incompatible links likely report contacts sampled by different substates within the broader ensemble.

This compatibility analysis helped explain why some conditions appeared similar by regional contact counts but separated by PCA of normalized intensities (Figure 6A). The regional contact-count analysis suggested similarity among pS129 dilute, WT 1.8 M TMAO, and pS129 1.8 M TMAO, but the high-intensity compatible links differed. For pS129 in dilute solution, the high normalized intensity of the K43–D121 cross-link supported a stronger long-range NT–CT contact (Figure 6B). In contrast, the 1.8 M TMAO samples were dominated by local NT contacts and had fewer high-intensity long-range cross-links. While the WT in 1.8 M TMAO had a compatible NT–CT cross-link, K45-E137, this cross-link involved a residue much closer to the C-terminus than for the pS129 dilute solution NT–CT cross-link in dilute solution. pS129 in 1.8 M TMAO lacked a compatible NT–CT cross-link. Thus, WT in 1.8 M TMAO can access broad contact classes similar to those accessed by pS129 dilute, but the dominant residue-level contacts are not the same.

Figure 6.

Figure 6.

Open in a new tab

A. Quantitative XL-MS workflow for resolving dominant contact patterns in IDP ensembles. Cross-links were first analyzed regionally to determine which broad conformer families were detected. Normalized cross-link intensities were then analyzed by PCA and k-means clustering to identify unique proteoform/environment-dependent residue-level contact patterns. High-intensity cross-links were next tested against STARLING-generated aSyn conformers to determine which restraints could be satisfied by the same individual conformer. B. pS129 dilute and WT 1.8 M TMAO share broad regional contact classes but differ in their long-range high-intensity compatible contacts. C. pS129 in 3.4 M TMAO and pS129 in OG share broad regional contact classes but differ in compatible high-intensity contacts, consistent with solution compaction in high TMAO and micelle-bound broken-helix-like remodeling for pS129 OG. The N-terminal region is dark blue, the NAC is black, and the C-terminal acidic region is red. Cross-links are shown on these example STARLING conformers as thick yellow lines.

A similar distinction was observed for WT 3.4 M TMAO, pS129 3.4 M TMAO, and pS129 OG. These conditions shared broad regional contact classes, but their compatible high-intensity contacts supported different conformational substate families. The 3.4 M TMAO conditions favored compact, solvent-driven substates containing the NT–CT K43-E114 cross-link and NT–NAC cross-links. In contrast, pS129 OG lacked high-intensity NT–CT cross-links, suggesting that the C-terminal tail is not collapsed onto the micelle-bound helical region (Figures 4E, 6C). Conditions can share broad regional contact classes but separate in PCA because their strongest compatible residue-level contacts are different.

These results also clarify how in vitro structural measurements should be interpreted in the context of cellular aSyn. The dilute condition provides a control state, but aSyn in cells experiences crowded cytosol,33 membrane surfaces,35 synaptic vesicle-associated interfaces,59 and disease-associated changes in metabolite composition.60,61 These high TMAO concentrations should not be interpreted as a direct mimic of intracellular concentration; rather, they provide a controlled osmolyte perturbation that favors compaction, and TMAO is a relevant choice for an osmolyte because it is a gut microbiome-derived metabolite linked to PD-associated metabolomic changes.40-43 In this context, the 1.8 and 3.4 M TMAO data show how solution compaction can reduce or override structural differences between WT and pS129 aSyn that are apparent in dilute buffer. OG micelles provide a complementary membrane-mimetic perturbation and show the opposite behavior: micelle binding separates WT and pS129 into distinct contact patterns. Therefore, cellular or disease-associated environments may not simply amplify the phosphorylation-induced compaction. Instead, they may reduce or change the structural effects of phosphorylation, depending on whether the biochemical environment imposes compaction or membrane binding.

CONCLUSIONS

These data show that S129 phosphorylation does not produce a fixed structural outcome for α-synuclein. Instead, phosphorylation changes how the aSyn conformational ensemble responds to its chemical environment. 1.8 M TMAO made WT aSyn more similar to dilute pS129 at the level of broad regional contacts, whereas 3.4 M TMAO collapsed both proteoforms and reduced the apparent proteoform-specific differences. In OG micelles, the two proteoforms diverged: WT showed an extended-helix-like contact pattern with stronger NT–CT contacts, while pS129 showed more NT–NAC contacts and fewer NT–CT restraints, consistent with a broken-helix-like topology and a dynamic C-terminal tail.

This conclusion depended on combining all three levels of analysis. Regional contact counts showed which types of contacts were detected under each condition, while PCA of normalized cross-link intensities separated conditions based on which residue-level contacts were most abundant. We then used a geometric compatibility analysis with STARLING-generated conformers to ask whether the strongest cross-link restraints could be satisfied by the same individual conformer. Together, these analyses show that similar regional contact patterns do not necessarily mean that the same parts of the ensemble dominate. More broadly, these data show that PTM effects in IDPs must be interpreted in the context of the chemical environment. S129 phosphorylation does not impose a single α-synuclein structure; instead, it changes how the ensemble responds to solution compaction and micelle binding. Quantitative XL-MS therefore provides a direct way to compare environment-dependent contact patterns in proteoform-defined IDP ensembles.

METHODS

See the Supporting Information (SI) for additional information about reagents, protein expression and purification, cross-linking reaction concentrations and times, trypsin and Glu-C digestion protocols, peptide quantitation methods, PCA with k-means clustering, and structural modeling.

Following quenching, cross-linked proteins were diluted to 5 μM protein with 250 mM ammonium acetate and then electrosprayed into either a Waters Synapt G2-Si IM-MS (Wilmslow, UK) equipped with an ExD cell (Agilent Technologies, Santa Clara, CA), Thermo Scientific Orbitrap Fusion Tribrid ETD Mass Spectrometer, or a Thermo Scientific Q Exactive Orbitrap Mass Spectrometer. Intact spectra were collected to determine the extent of cross-linking. The experimental workflow is shown in Figure S1.

A Waters ACQUITY UPLC M-Class coupled online to the Fusion MS was used for the bottom-up experiments. Peptide mixtures were diluted to ~ 70 ng/μL in 250 mM ammonium acetate pH 7. 70 ng of XL peptides were injected by a Waters ACQUITY UPLC M-Class via an in-house packed trap column coupled to a C18 analytical column. Columns were packed using a Next Advance pressure injection cell with a magnetic stir plate using Dr. Maisch HPLC GmbH Reprosil-Pur 120 C18-AQ, 5 μm packing material, and LC grade methanol. The trap column was ~ 5 cm long and ~ 175 μm ID. The analytical column was ~ 30 cm long and ~ 75 μm ID. Solvent A was 95:5 water/acetonitrile with 0.1% formic acid, and solvent B was acetonitrile with 0.1% formic acid. The sample was loaded at 95% A onto the trapping column for 5 min at 5 μL/min, and then the flow rate was slowed to 0.3 μL/min. After loading, a 95% to 40% A gradient was applied over 90 min to elute peptides to the MS.

Experiments were conducted in positive mode. The Thermo Scientific Orbitrap Fusion MS settings were an AGC target of 3e5, a maximum injection time of 50 ms, and a m/z range of 350 to 3500 in high mass mode at 120,000 resolution. The dd-MS2 settings were set to have a resolution of 60,000, AGC target of 3e5, maximum injection time of 118 ms, data-dependent acquisition with the Top Speed 3 s method, a precursor isolation window 1.6 m/z, and the m/z range set to auto. Assisted HCD was set at 15, 25, and 35 V. A dynamic exclusion time of 7 s was used after a precursor was selected three times. The data for the bottom-up approach were analyzed for three replicates for each of the XL reaction mixtures. To annotate and view cross-linked and monolinked MS and MS/MS spectra, pLink/pLabel were used with a false discovery rate (FDR) of 1%.62 Cross-linking maps were constructed with xiNET.63

For quantitative PCA, cross-linked peptides identified in at least two of three replicates were quantified from extracted ion chromatograms in Skyline64 and normalized within each condition as described in the Supporting Information. The resulting cross-link intensity matrix contained residue pairs as variables and replicate peptide intensities as observations. PCA was performed on the normalized intensity matrix to compare the cross-linking patterns across proteoforms and environments. The number of principal components retained was selected from an elbow plot, and k-means clustering was used to group similar quantitative cross-linking profiles with the number of clusters selected by silhouette analysis. Regional cross-linking count analysis and PCA were intentionally treated as separate levels of analysis: regional counts identify which broad conformer families are detected, whereas PCA identifies which residue-level contacts dominate.

For cross-link compatibility analysis, the highest-intensity normalized cross-links from each condition were mapped onto 10,000 aSyn conformers generated with STARLING. STARLING was used only to provide a diverse set of conformers. Each cross-link was tested by using the distance threshold for its corresponding chemistry, and compatible contact sets were defined as high-intensity restraints that could be satisfied by the same individual conformer. This analysis was used as a qualitative test of whether dominant residue-level contacts could coexist within a conformational subfamily.

Supplementary Material

Supporting Information
Contact analysis
PCA Analysis

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacs.6c10987.

  • Additional information about reagents, protein expression and purification, cross-linking reaction conditions, trypsin and Glu-C digestion, cross-linked peptide quantitation, PCA with k-means clustering, and structural modeling (PDF)

  • Final PCA/k-means clustering analyses (XLSX)

  • Chi-squared tests for independence (XLSX)

ACKNOWLEDGMENTS

This work was funded by the National Institute of General Medical Sciences of the National Institutes of Health under R35GM151251 (IKW).

Footnotes

The authors declare no competing financial interest.

Contributor Information

Ashlyn N. Dollar, Department of Chemistry and Chemical Biology, Indiana University Indianapolis, Indianapolis, Indiana 46202, United States

Daniel W. Kelley, Department of Chemistry and Chemical Biology, Indiana University Indianapolis, Indianapolis, Indiana 46202, United States

Ian K. Webb, Department of Chemistry and Chemical Biology, Indiana University Indianapolis, Indianapolis, Indiana 46202, United States; Center for Computational Biology and Bioinformatics, Indiana University School of Medicine, Indianapolis, Indiana 46202, United States.

Data Availability Statement

LC-MS/MS data sets and intact cross-linked protein mass spectra are available on MassIVE with data set identifier MSV000102384.

REFERENCES

  • (1).Uversky VN. Protein intrinsic disorder and structure-function continuum. Progress in Molecular Biology and Translational Science, Uversky VN, Ed.; Academic Press, 2019; Vol. 166, pp 1–17. [DOI] [PubMed] [Google Scholar]
  • (2).Bondos SE; Dunker AK; Uversky VN. On the roles of intrinsically disordered proteins and regions in cell communication and signaling. Cell Commun. Signal. 2021, 19 (1), 88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (3).Trivedi R; Nagarajaram HA. Intrinsically Disordered Proteins: An Overview. Int. J. Mol. Sci. 2022, 23 (22), 14050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (4).Sulzer D; Edwards RH. The physiological role of α-synuclein and its relationship to Parkinson’s Disease. J. Neurochem. 2019, 150 (5), 475–486. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (5).Calabresi P; Mechelli A; Natale G; Volpicelli-Daley L; Di Lazzaro G; Ghiglieri V. Alpha-synuclein in Parkinson’s disease and other synucleinopathies: from overt neurodegeneration back to early synaptic dysfunction. Cell Death Dis. 2023, 14 (3), 176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (6).Stefanis L. α- Synuclein in Parkinson’s disease. Cold Spring Harbor Pesrpect. Med. 2012, 2 (2), a009399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (7).Burré J; Sharma M; Südhof TC. α-Synuclein assembles into higher-order multimers upon membrane binding to promote SNARE complex formation. Proc. Natl. Acad. Sci. U.S.A. 2014, 111 (40), E4274–4283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (8).Emamzadeh FN. Alpha-synuclein structure, functions, and interactions. J. Res. Med. Sci. 2016, 21, 29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (9).Gallardo J; Escalona-Noguero C; Sot B. Role of α-Synuclein Regions in Nucleation and Elongation of Amyloid Fiber Assembly. ACS Chem. Neurosci. 2020, 11 (6), 872–879. [DOI] [PubMed] [Google Scholar]
  • (10).Kawahata I; Finkelstein DI; Fukunaga K. Pathogenic Impact of α-Synuclein Phosphorylation and Its Kinases in α-Synucleinopathies. Int. J. Mol. Sci. 2022, 23 (11), 6216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (11).Sato H; Kato T; Arawaka S. The role of Ser129 phosphorylation of α-synuclein in neurodegeneration of Parkinson’s disease: a review of in vivo models. Rev. Neurosci. 2013, 24 (2), 115–123. [DOI] [PubMed] [Google Scholar]
  • (12).Fujiwara H; Hasegawa M; Dohmae N; Kawashima A; Masliah E; Goldberg MS; Shen J; Takio K; Iwatsubo T. α-Synuclein is phosphorylated in synucleinopathy lesions. Nat. Cell Biol. 2002, 4 (2), 160–164. [DOI] [PubMed] [Google Scholar]
  • (13).Ramalingam N; Jin S-X; Moors TE; Fonseca-Ornelas L; Shimanaka K; Lei S; Cam HP; Watson AH; Brontesi L; Ding L; et al. Dynamic physiological α-synuclein S129 phosphorylation is driven by neuronal activity. npj Parkinson's Disease 2023, 9 (1), 4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (14).Ramalingam N; Haass C; Dettmer U. Physiological roles of α-synuclein serine-129 phosphorylation - not an oxymoron. Trends Neurosci. 2024, 47 (7), 480–490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (15).Weston LJ; Cook ZT; Stackhouse TL; Sal MK; Schultz BI; Tobias ZJC; Osterberg VR; Brockway NL; Pizano S; Glover G; et al. In vivo aggregation of presynaptic alpha-synuclein is not influenced by its phosphorylation at serine-129. Neurobiology of Disease 2021, 152, No. 105291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (16).Kuwahara T; Tonegawa R; Ito G; Mitani S; Iwatsubo T. Phosphorylation of α-Synuclein Protein at Ser-129 Reduces Neuronal Dysfunction by Lowering Its Membrane Binding Property in Caenorhabditis elegans. J. Biol. Chem. 2012, 287 (10), 7098–7109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (17).Waxman EA; Giasson BI. Characterization of kinases involved in the phosphorylation of aggregated α-synuclein. J. Neurosci. Res. 2011, 89 (2), 231–247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (18).Paleologou KE; Schmid AW; Rospigliosi CC; Kim H-Y; Lamberto GR; Fredenburg RA; Lansbury PT Jr.; Fernandez CO; Eliezer D; Zweckstetter M; et al. Phosphorylation at Ser-129 but Not the Phosphomimics S129E/D Inhibits the Fibrillation of α-Synuclein. J. Biol. Chem. 2008, 283 (24), 16895–16905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (19).Chen L; Feany MB. α-Synuclein phosphorylation controls neurotoxicity and inclusion formation in a Drosophila model of Parkinson disease. Nat. Neurosci. 2005, 8 (5), 657–663. [DOI] [PubMed] [Google Scholar]
  • (20).Wu B; Liu Q; Duan C; Li Y; Yu S; Chan P; Uéda K; Yang H. Phosphorylation of α-synuclein upregulates tyrosine hydroxylase activity in MN9D cells. Acta Histochemica 2011, 113 (1), 32–35. [DOI] [PubMed] [Google Scholar]
  • (21).Gorbatyuk OS; Li S; Sullivan LF; Chen W; Kondrikova G; Manfredsson FP; Mandel RJ; Muzyczka N. The phosphorylation state of Ser-129 in human α-synuclein determines neurodegeneration in a rat model of Parkinson disease. Proc. Natl. Acad. Sci. U.S.A. 2008, 105 (2), 763–768. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (22).Arawaka S; Wada M; Goto S; Karube H; Sakamoto M; Ren C-H; Koyama S; Nagasawa H; Kimura H; Kawanami T; et al. The Role of G-Protein-Coupled Receptor Kinase 5 in Pathogenesis of Sporadic Parkinson’s Disease. J. Neurosci. 2006, 26 (36), 9227–9238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (23).Freichel C; Neumann M; Ballard T; Müller V; Woolley M; Ozmen L; Borroni E; Kretzschmar HA; Haass C; Spooren W; et al. Age-dependent cognitive decline and amygdala pathology in α-synuclein transgenic mice. Neurobiol. Aging 2007, 28 (9), 1421–1435. [DOI] [PubMed] [Google Scholar]
  • (24).Kahle PJ; Neumann M; Ozmen L; Müller V; Jacobsen H; Spooren W; Fuss B; Mallon B; Macklin WB; Fujiwara H; et al. Hyperphosphorylation and insolubility of α-synuclein in transgenic mouse oligodendrocytes. EMBO Reports 2002, 3 (6), 583–588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (25).Luk KC; Song C; O’Brien P; Stieber A; Branch JR; Brunden KR; Trojanowski JQ; Lee VM-Y. Exogenous α-synuclein fibrils seed the formation of Lewy body-like intracellular inclusions in cultured cells. Proc. Natl. Acad. Sci. U.S.A. 2009, 106 (47), 20051–20056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (26).Volpicelli-Daley LA; Luk KC; Patel TP; Tanik SA; Riddle DM; Stieber A; Meaney DF; Trojanowski JQ; Lee VM-Y. Exogenous α-Synuclein Fibrils Induce Lewy Body Pathology Leading to Synaptic Dysfunction and Neuron Death. Neuron 2011, 72 (1), 57–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (27).de Bruyn E; Dorn AE; Rossetti G; Fernandez C; Outeiro TF; Schulz JB; Carloni P. Impact of Phosphorylation on the Physiological Form of Human alpha-Synuclein in Aqueous Solution. J. Chem. Inf. Model. 2024, 64 (21), 8215–8226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (28).Ma M-R; Hu Z-W; Zhao Y-F; Chen Y-X; Li Y-M. Phosphorylation induces distinct alpha-synuclein strain formation. Sci. Rep. 2016, 6 (1), No. 37130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (29).Parra-Rivas LA; Madhivanan K; Aulston BD; Wang L; Prakashchand DD; Boyer NP; Saia-Cereda VM; Branes-Guerrero K; Pizzo DP; Bagchi P; et al. Serine-129 phosphorylation of α-synuclein is an activity-dependent trigger for physiologic protein-protein interactions and synaptic function. Neuron 2023, 111 (24), 4006–4023.e10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (30).Dent SE; King DP; Osterberg VR; Adams EK; Mackiewicz MR; Weissman TA; Unni VK. Phosphorylation of the aggregate-forming protein alpha-synuclein on serine-129 inhibits its DNA-bending properties. J. Biol. Chem. 2022, 298 (2), 101552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (31).Semenyuk PI. Alpha-synuclein phosphorylation induces amyloid conversion via enhanced electrostatic bridging: Insights from molecular modeling of the full-length protein. Biophys. Chem. 2024, 307, No. 107196. [DOI] [PubMed] [Google Scholar]
  • (32).Dollar AN; Webb IK. Cross-Linking and Covalent Labeling Mass Spectrometry Reveal Proteoform-Driven Conformational Changes in Alpha Synuclein. Anal. Chem. 2026, 98 (4), 2765–2774. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (33).Theillet F-X; Binolfi A; Bekei B; Martorana A; Rose HM; Stuiver M; Verzini S; Lorenz D; van Rossum M; Goldfarb D; et al. Structural disorder of monomeric α-synuclein persists in mammalian cells. Nature 2016, 530 (7588), 45–50. [DOI] [PubMed] [Google Scholar]
  • (34).Fisher SK; Cheema TA; Foster DJ; Heacock AM. Volume-dependent osmolyte efflux from neural tissues: regulation by G-protein-coupled receptors. J. Neurochem. 2008, 106 (5), 1998–2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (35).Kaur U; Lee JC. Unroofing site-specific α-synuclein-lipid interactions at the plasma membrane. Proc. Natl. Acad. Sci. U.S.A. 2020, 117 (32), 18977–18983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (36).Pincus DL; Hyeon C; Thirumalai D. Effects of trimethylamine N-oxide (TMAO) and crowding agents on the stability of RNA hairpins. J. Am. Chem. Soc. 2008, 130 (23), 7364–7372. [DOI] [PubMed] [Google Scholar]
  • (37).Ma J; Pazos IM; Gai F. Microscopic insights into the protein-stabilizing effect of trimethylamine N-oxide (TMAO). Proc. Natl. Acad. Sci. U.S.A. 2014, 111 (23), 8476–8481. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (38).Sung H-L; Sengupta A; Nesbitt D. Smaller molecules crowd better: Crowder size dependence revealed by single-molecule FRET studies and depletion force modeling analysis. J. Chem. Phys. 2021, 154 (15), DOI: 10.1063/5.0045492. [DOI] [PubMed] [Google Scholar]
  • (39).Hilaire MR; Abaskharon RM; Gai F. Biomolecular Crowding Arising from Small Molecules, Molecular Constraints, Surface Packing, and Nano-Confinement. J. Phys. Chem. Lett. 2015, 6 (13), 2546–2553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (40).Morais LH; Boktor JC; MahmoudianDehkordi S; Kaddurah-Daouk R; Mazmanian SK. α-synuclein overexpression and the microbiome shape the gut and brain metabolome in mice. npj Parkinson's Disease 2024, 10 (1), 208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (41).Loh JS; Mak WQ; Tan LKS; Ng CX; Chan HH; Yeow SH; Foo JB; Ong YS; How CW; Khaw KY. Microbiota–gut–brain axis and its therapeutic applications in neurodegenerative diseases. Signal Transduction and Targeted Therapy 2024, 9 (1), 37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (42).Chen S-J; Kuo C-H; Kuo H-C; Chen C-C; Wu W-K; Liou J-M; Wu M-S; Lin C-H. The Gut Metabolite Trimethylamine N-oxide Is Associated With Parkinson’s Disease Severity and Progression. Movement Disorders 2020, 35 (11), 2115–2116. [DOI] [PubMed] [Google Scholar]
  • (43).Chung SJ; Rim JH; Ji D; Lee S; Yoo HS; Jung JH; Baik K; Choi Y; Ye BS; Sohn YH; et al. Gut microbiota-derived metabolite trimethylamine N-oxide as a biomarker in early Parkinson’s disease. Nutrition 2021, 83, No. 111090. [DOI] [PubMed] [Google Scholar]
  • (44).Uversky VN; Li J; Fink AL. Trimethylamine-N-oxide-induced folding of alpha-synuclein. FEBS Lett. 2001, 509 (1), 31–35. [DOI] [PubMed] [Google Scholar]
  • (45).Jamal S; Kumari A; Singh A; Goyal S; Grover A. Conformational Ensembles of α-Synuclein Derived Peptide with Different Osmolytes from Temperature Replica Exchange Sampling. Front. Neurosci. 2017, 11, 684. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (46).Meade RM; Allen SG; Williams C; Tang TMS; Crump MP; Mason JM. An N-terminal alpha-synuclein fragment binds lipid vesicles to modulate lipid-induced aggregation. Cell Reports Physical Science 2023, 4 (9), 101563. [Google Scholar]
  • (47).Sarchione A; Marchand A; Taymans JM; Chartier-Harlin MC. Alpha-Synuclein and Lipids: The Elephant in the Room? Cells 2021, 10 (9), 2452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (48).Ulmer TS; Bax A; Cole NB; Nussbaum RL. Structure and Dynamics of Micelle-bound Human α-Synuclein. J. Biol. Chem. 2005, 280 (10), 9595–9603. [DOI] [PubMed] [Google Scholar]
  • (49).Georgieva ER; Ramlall TF; Borbat PP; Freed JH; Eliezer D. The Lipid-binding Domain of Wild Type and Mutant α-Synuclein: Compactness and Interconversion Between the Broken and Extended Helix Forms. J. Biol. Chem. 2010, 285 (36), 28261–28274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (50).Wang C; Zhang K; Cai B; Haller JE; Carnazza KE; Hu J; Zhao C; Tian Z; Hu X; Hall D; et al. VAMP2 chaperones α-synuclein in synaptic vesicle co-condensates. Nat. Cell Biol. 2024, 26 (8), 1287–1295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (51).Dollar AN; Webb IK. Strategies for Acid and Amine Cross-linking and Labeling for Protein Structural Characterization Using Mass Spectrometry. Anal. Chem. 2025, 97 (15), 8213–8219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (52).Sriswasdi S; Harper SL; Tang HY; Speicher DW. Enhanced identification of zero-length chemical cross-links using label-free quantitation and high-resolution fragment ion spectra. J. Proteome Res. 2014, 13 (2), 898–914. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (53).Piersimoni L; Kastritis PL; Arlt C; Sinz A. Cross-Linking Mass Spectrometry for Investigating Protein Conformations and Protein–Protein Interactions—A Method for All Seasons. Chem. Rev. 2022, 122 (8), 7500–7531. [DOI] [PubMed] [Google Scholar]
  • (54).Ubbiali D; Fratini M; Piersimoni L; Ihling CH; Kipping M; Heilmann I; Iacobucci C; Sinz A. Direct Observation of ″Elongated″ Conformational States in α-Synuclein upon Liquid-Liquid Phase Separation. Angew. Chem., Int. Ed. 2022, 61 (46), No. e202205726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (55).Sarma R; Paul S. Interactions of S-peptide analogue in aqueous urea and trimethylamine-N-oxide solutions: A molecular dynamics simulation study. J. Chem. Phys. 2013, 139 (3), DOI: 10.1063/1.4813502. [DOI] [PubMed] [Google Scholar]
  • (56).Kameyama K; Takagi T. Micellar properties of octylglucoside in aqueous solutions. J. Colloid Interface Sci. 1990, 137 (1), 1–10. [Google Scholar]
  • (57).Lokappa SB; Ulmer TS. α-Synuclein Populates Both Elongated and Broken Helix States on Small Unilamellar Vesicles. J. Biol. Chem. 2011, 286 (24), 21450–21457. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (58).Novak B; Lotthammer JM; Emenecker RJ; Holehouse AS. Accurate predictions of disordered protein ensembles with STARLING. Nature 2026, 652 (8108), 240–250. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (59).Gao V; Briano JA; Komer LE; Burré J. Functional and Pathological Effects of α-Synuclein on Synaptic SNARE Complexes. J. Mol. Biol. 2023, 435 (1), No. 167714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (60).Perovnik M; Tomše P; Jamšek J; Tang C; Eidelberg D; Trošt M. Metabolic brain pattern in dementia with Lewy bodies: Relationship to Alzheimer’s disease topography. Neuroimage Clin 2022, 35, No. 103080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (61).Anandhan A; Jacome MS; Lei S; Hernandez-Franco P; Pappa A; Panayiotidis MI; Powers R; Franco R. Metabolic Dysfunction in Parkinson’s Disease: Bioenergetics, Redox Homeostasis and Central Carbon Metabolism. Brain Res. Bull. 2017, 133, 12–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (62).Chen Z-L; Meng J-M; Cao Y; Yin J-L; Fang R-Q; Fan S-B; Liu C; Zeng W-F; Ding Y-H; Tan D; et al. A high-speed search engine pLink 2 with systematic evaluation for proteome-scale identification of cross-linked peptides. Nat. Commun. 2019, 10 (1), 3404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (63).Combe CW; Fischer L; Rappsilber J. xiNET: cross-link network maps with residue resolution. Mol. Cell Proteomics 2015, 14 (4), 1137–1147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (64).MacLean B; Tomazela DM; Shulman N; Chambers M; Finney GL; Frewen B; Kern R; Tabb DL; Liebler DC; MacCoss MJ. Skyline: an open source document editor for creating and analyzing targeted proteomics experiments. Bioinformatics 2010, 26 (7), 966–968. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
NIHMS2207809-supplement-Supporting_Information.pdf (264.6KB, pdf)
Contact analysis
NIHMS2207809-supplement-Contact_analysis.xlsx (131.6KB, xlsx)
PCA Analysis
NIHMS2207809-supplement-PCA_Analysis.xlsx (192.7KB, xlsx)

Data Availability Statement

LC-MS/MS data sets and intact cross-linked protein mass spectra are available on MassIVE with data set identifier MSV000102384.

RESOURCES