Conformational Dynamics of α-Synuclein: A Study of its Intramolecular Forces in the Presence of Selected Compounds

doi:10.21203/rs.3.rs-3137135/v1

Download PDF

Article

Conformational Dynamics of α-Synuclein: A Study of its Intramolecular Forces in the Presence of Selected Compounds

https://doi.org/10.21203/rs.3.rs-3137135/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 03 Nov, 2023

Read the published version in Scientific Reports →

You are reading this latest preprint version

Protein misfolding and aggregation play a role in amyloidogenic diseases through the self-assembly of intrinsically disordered proteins (IDPs) in type II diabetes (T2D), Alzheimer's (AD) and Parkinson's (PD) diseases. PD is the most common neurodegenerative disorder after AD, known for the loss of dopaminergic signaling, which causes motor and non-motor signs and symptoms. Lewy bodies and Lewy neurites are common pathological hallmarks of PD that are mainly composed of an aggregate of the disordered protein, α-synuclein (α-Syn). There have been many efforts to develop chemical-based compounds to prevent aggregation or facilitate disruption of the fibrils. These have been tested in wet labs, but most fail to generate a robust impact. Further, the atomistic roles and interactions of such compounds have yet to be revealed. The conformational diversity and detailed interactions among homo-oligomer chains of α-Syn are also unknown; identifying these might help uncover a practical approach to developing a potent therapy. In this study, we use an in-silico investigation to address the conformational diversity of α-Syn oligomers. The roles of several point mutations in protein aggregation in PD are known; we take this further by evaluating the interactional energies and contributions of all residues in stability and chain–chain interactions. We dock three chemical derivatives of known compounds with high-score drug-likeness to evaluate the roles of our ligands in the conformational dynamicity of the oligomers, with emphasis on intramolecular forces. Preventing fibril formations is a heated topic in this area. Free energy evaluation of the modeled inter- and intramolecular interactions through MD simulation shows strong binding between α-Syn compounds. However, we find that they do not disrupt or even weaken the interactions, and in some cases, they contributed to boosting interactions between oligomer chains.

Biological sciences/Biophysics

Biological sciences/Computational biology and bioinformatics

Biological sciences/Drug discovery

Biological sciences/Genetics

Biological sciences/Structural biology

Up to 3% of the global population over 65 years old are affected by Parkinson's Disease (PD), the second leading neurodegenerative disorder after Alzheimer's disease (AD). PD is a movement disorder of the dopaminergic neurons from the substantia nigra (1, 2). A well-known misfolded protein in PD is α-synuclein (α-Syn), the main constituent of aggregates that form Lewy bodies (LB). The term "synuclein" refers to its synapse and nuclear envelope localization. Lewy bodies are major neuropathological characteristics of PD and dementia, α-Syn is the critical component that forms pale bodies and Lewy neurites (LN) (3–5). Lewy bodies are mainly cortical or brainstem (classical), with up to 30 µm in diameter of spherical inclusions (6). α-Syn aggregates appear either as plaque for extracellular deposition or inclusion for intracellular bodies, commonly observed in amyloid disorders such as PD (7). Decades of studies have examined the conformational changes, aggregates, and testing of α-Syn in search of ways to stop, lower the rate, or reverse PD aggregation (8–10). α-Syn controls dopaminergic neurotransmission and is involved in soluble N-ethylmaleimide-sensitive factor-attachment protein receptor (SNARE-complex) assembly (10–12), although the function for this has yet to be explained fully. Generally, any clinical condition initiated (thermodynamically) via inappropriate protein folding with a high β-sheet content and fibrillar characteristics compared to its natural soluble form is called amyloidosis (10, 13, 14). The inclusion of α-Syn, mainly seen in LB and LN, is called synucleinopathy; this is abundant with β-sheet aggregates (8). α-Syn is a SNCA gene product, in humans, it is expressed in neurons' presynaptic terminals, composed of 140 amino acids (aa). It has three segments of N-ter for 60 aa, a non-amyloid-β component, which is hydrophobic primarily and composed of amino acids 61 to 95 to connect the N-ter and the acidic C-ter, from aa 96 to 140 (15, 16). The lipid binding domain is located within the first 100 aa at N-ter, compose of seven non-perfect repeats of 11 aa. The main domain for its aggregation is the hydrophobic domain which is considered for investigation in this study (17). α-Syn is very abundant in the brain, with ~ 1% of total rat brain proteins being α-Syn. It is also found in the heart, lungs, muscles, kidneys, and red blood cells (11). Available treatments, such as cholinesterase, dopamine agonist, and monoamine oxidase, have not been fully effective; further, their potential decreases during disease progression (18, 19). Opposite to soluble and highly disordered monomer (yet functional) α-Syn in healthy individuals, oligomers, protofibrils, and LB aggregates are highly organized, ordered, and insoluble. These line up into compact β-sheet structures (20). Studies to date have mainly addressed the α-syn monomer and its process of aggregation, potential chemicals for preventing aggregation, and its conformational plasticity. However, the exact role and interactions of each deformed and conformationally modified chains of α-syn remain uncertain with respect to each other and whole oligomers. Studying the dynamics of these chains and whole oligomers might yield useful information to better design chemical compounds or even monoclonal antibodies against aggregates (21–24). The size of an α-Syn protofibrils is around 7 nm, ~ 3 nm less than mature fibrils, which in vitro measurements reveal as roughly 10 nm (25). Encountering aggregate generation is impossible without adequate knowledge of the atomistic and large-scale structural motions of the forming atoms and their interactions with other aggregate components. Any small molecule, peptide, or antibody that reduces the aggregation rate or affects oligomer conformation to induce the generation of non-pathogenic species may be considered a possible strategy for tackling the problem. In this study, we select the most ordered core protein in an oligomer arrangement with six chains. We observe all motions of these six chains without any constraint to evaluate their flexibility, the role of the chemical compounds that may reduce or even increase flexibility. We seek to find the most energetic and weak interaction regions, identify the effective residues in weak or strong regions, and determine the most effective target site for designing chemical compounds or nanobodies to weaken their interactions. The helix conformation of α-Syn is mostly observed when it is in its functional form and interacting with vesicles and membranes; otherwise, it undergoes conformational changes to lose the necessary interactions required for helix stabilization in favor of the β-sheet strand, aggregation, fibril and at the end LB formation (11). The exact atomistic details of protein inclusion and α-syn aggregation are not yet fully understood. However, some studies show that LB formation begins with the monomeric form of α-Syn (i.e., seeds), which extend into twisted amyloid fibrils before developing fully into mature fibrils. The seeds are composed of pathogenic β-sheet-rich fibrils (11, 26). This transformation or conformational change leads to neuronal death because it disrupts almost all α-Syn biological functions, including vesicular transport and organelle functions. Studies estimate the number of oligomer chains required to initiate the larger aggregate, ensure stability, and trigger the seeding capacity to > 15 (26). Thus, we aim to study the dynamics of oligomers with fewer than 15 chains to uncover what occurs when fewer chains interact with each other or with a chemical compound. Different levels of neurotoxicity relate to soluble and insoluble α-Syn species (oligomers, prefibrillar, and fully mature fibrils). All these species can be toxic, but the extent to which one disrupts cellular homeostasis and leads to neuronal death is under debate. Some studies indicate that mature fibrils are highly toxic due to their ability to interact and permeabilize cell membranes (27, 28). Other assessments assert that oligomers as the more toxic species (29–31).

3.1. Docking Simulation

Three chemical compounds have been selected for study of their interactions and impact on the stability of α-Syn. These are C14, or (4-((R)-2-((R)-2,2,2-trifluoro-1-hydroxyethyl)pyrrolidin-1-yl)-2-trifluoromethyl) benzonitrile; Gingerol, or (5S)-5-hydroxy-1-(4-hydroxy-3-methoxyphenyl)decan-3-one (Gin)); and C10, or 4-(Trifluoromethyl) cinnamaldehyde.

Pujols et al. (32) report that small lipophilic molecules with trifluoromethyl on their structures are promising candidates for PD treatment. For this reason, we screen similar compounds and select two synthetic ones (C10 and C14) for molecular dynamics simulations, in addition to one naturally occurring compound. There is much experimental evidence of ginger's beneficial effects on PD development (33). Thus, we consider gingerol (the main pharmaceutically active compound in ginger) in our simulations.

To model the interaction between ligands (4-(®-2-((R)-2,2,2-trifluoro-1-hydroxyethpyrrolidinedin-1-yl)-2-trifluoromethyl) benzonitrile (C14) and (5S)-5-hydroxy-1-(4-hydroxy-3-methoxyphenyl) decan-3-one (Gin)), 4-(Trifluoromethyl) cinnamaldehyde (C10), and the α-Syn fiber, we obtained the cryo-electron microscopy structure of the α-Syn fiber (PDB: 6A6B) from the Protein Data Bank, (http://www.pdb.org ), as a receptor for the docking and MD simulations. The geometries of the ligands are shown in Fig. 1a-c. iGemdock v2. 1 docking tool was used to find the most energetic poses between compounds and the α-Syn (34). This tool evaluates the hydrogen-bonding (H), Van der Waal's (V), and electrostatic (E) interactions. We estimated the empirical scoring function of iGemdock as follows: binding energy = vdW + Hbond + Elec (Fig. 1a-c). The calculated binding energy for C14 and α-Syn fiber was − 72.4 in a unit of the software's score, including the contributions of hydrogen and Van der Waal's bonding − 10.0 and − 62.40, respectively. Similar parameters were calculated for gingerol (Gin) and C10, the binding energy was − 90.24, and the contributions of hydrogen and Van der Waal's were − 19.32 and − 70.92, respectively for Gin. For C10, the binding energy was − 71.72, the Van der Waal's contribution for this complex was − 65, and the hydrogen contribution was − 6.71.

3.2. Molecular Dynamics Simulation

The Molecular Dynamics simulation (MDs) of the oligomer structure of α-syn in interactions with three chemical compounds and one unliganded simulation was performed for 300 nano-second (ns) using 2021.1 GROMACS (35, 36), and GROMOS96 54A7 force field (37, 38). The GROMOS96 G54A7FF United-Atom topologies (ITP file) for all three compounds were obtained from Automated Topology Builder (ATB) and Repository (39). A dodecahedron box with periodic boundary conditions and 1.0 nm distances between the box edge and complexes was defined before SPC water models filled it out. Systems were neutralized by adding Na⁺ atoms and were energy minimized by the steepest descent minimization algorithm. To avoid atomic clashes before starting the simulation, and after minimization, all conformations were compared with the initial coordinates. The time step for each simulation was set to 2 fs, and the "md" integrator was employed to integrate Newton's equations of motion. The electrostatic (long-range) and van der Waals (short-range) interactions were treated by Particle Mesh Ewald (40) and Lennard Jones, respectively, while applying a 1.2 nm cut-off. The temperature and pressure were kept stable at 300 k and 1 bar by assigning a modified Berendsen thermostat (41, 42) (V-rescale) applying time constant of τ_t = 0.1 ps and Parrinello-Rahman pressure coupling with the compressibility of 4.5e-5 (43). All simulations used the LINCS algorithm as the constraint algorithm for bond length. Before the production simulation for 300 ns the Position Restrain (PR, known as NVT ensemble with constant Number of particles, Volume, and Temperature)) and NPT simulation (NPT ensemble which is constant for Number of particles, Pressure, and Temperature)) were conducted for one ns and five ns, respectively, with the same 2 fs time step. The five ns of NPT simulation was performed after the NVT simulation. PyMol (https://pymol.org/2/), Grace (https://plasma-gate.weizmann.ac.il/Grace/ ) and VMD (https://www.ks.uiuc.edu/Research/vmd/), were used for visualization or generating graphs.

3.3. Free Energy Evaluation

Each compound's free binding energy was calculated through the g_mmpbsa tool from 250 ns – 300 ns with - dt set to 100 ps. The g_mmpbsa is applying molecular mechanics Poisson–Boltzmann surface area (MM-PBSA) to evaluate free energies (44, 45).

The binding free energy is measured as

∆G_binding = G_{α−Syn − Compounds} – (G_α−Syn + G _Compounds) Eq: 1

The total energy between α-Syn and ligands can be seen as “G_{α−Syn – Compounds”}. The free energy, in general, for each entity was evaluated according to Eq. 2 (44).

G_L =\(⟨{E}_{MM}⟩\) - TS + \(⟨{G}_{Solvation}⟩\) Eq: 2

The L indicates α-Syn-Compounds, and 〈E_MM 〉 measures average vacuum potential energy, which is the potential energy for bonded and nonbonded interactions. The nonbonded interactions of electrostatic, and van der Waals were calculated using the g_mmpbsa tool as well. The bonded interaction measures the contribution of improper, angles, bonds, and dihedrals. The entropy (S) and temperature (T) are evaluated as the TS in Eq. 2 and refers to entropic contribution to free energy in a vacuum. The average free energy of solvation as it requires to move solute from a vacuum into the solvent counts as \(⟨{G}_{Solvation}⟩\). Notably, the g_mmpbsa solvent is not explicit, and the free energy of solvation includes polar and nonpolar solvents (44).

To see the stability of each chain in different simulation the Root Mean Square deviation (RMSD) was calculated. The reference structure for comparison was set to the widely accepted minimized equilibrated conformation. The backbone atoms of chains 1 and 6 (belonging to an unliganded oligomer) showed a higher value of RMS deviation for chain 1 (~ 10 Å) and a none-converged value of ~ 5.5 to 8 Å from 155 ns for chain 6 (Fig. 2a & c). It was unclear whether this could be attributed to more dimensional freedom or was the result of chemical interactions. We did not intend to claim whether our compounds would have any therapeutic impact, but rather sought to study their interactions on the chain's stability. Compounds were selected if they or their root compounds previously showed some impact on α-Syn or other proteins; all those selected had strong drug-likeness parameters, as revealed through the evaluation of Lipinski's "rule of five" (Fig. 1d).

The RMSDs for chains 1 and 2 of the three simulations (Gin, C10, and C14) showed convergence and stable behavior from 2 to a maximum of 5 Å, which consider as an acceptable range. The lowest RMSD for chain 2 was seen for C14, from 150 to 300 ns at the range of 2 to 3 Å. Chains 3 and 4 for the unliganded and all liganded simulations fluctuated in the range of 2 to 4 Å, larger RMSDs were observed for both chains in the unliganded simulation. The unliganded chain 5 (with ~ 5 Å) and chain 6 of the C14 simulations were less stable than chains 5 and 6 in the C10 and Gin simulations. Chains 1 and 6 were the most unstable, thus, we further investigated the behaviors of these two chains in addition to four chains with liganded and unliganded simulations. Most chains showed convergence based on their RMSD assessments, apart from chain 6 of C14 during the last 50 ns (Fig. 2a-c).

Apart from the docking analysis result, to ensure strong interaction between chemical compounds and α-Syn, we evaluate free energy through a detailed explanation of the material and methods (Section 3.3 and Fig. 2d). The result confirmed that Gin makes more energetic binding to the oligomer with − 58.3 kcal/mol compared to -38.9 kcal/mol in C14 and − 25.6 kcal/mol in C10 (Fig. 2d). This result, combined with the docking scores and "Lipinski's rule of five" assessment indicating strong binding between selected compounds and α-Syn oligomers.

We then evaluated hydrogen bonds at the cut-off of 3.5 Å between each neighboring chain to assess the H-bond numbers and insights into the roles of chemical compounds on chain instability. The number of H-bonds between chains 1 and 2 and between chains 2 and 3 in the unliganded system fluctuated between ~ 37 and ~ 44, while for all liganded simulations, they ranged from a minimum of ~ 52 to a maximum of ~ 62. We observed the same trend for H-bonds between chains 3 and 4 and between chains 4 and 5, as these ranged from ~ 43 to ~ 52 in the unliganded simulation and 54 to 64 in the liganded simulation. From these observations, we conclude that each chemical compound may contribute to more potent or effective interactions among protofibril chains and that the three applied chemicals may stabilize them (Fig. 3a - b). The number of H-bonds between chains 5 and 6 in the unliganded simulation was 25 to 40 compared to the average of 50 to 69 in all other simulations and other chains (Supp Fig. 1).

By considering the RMSD and number of H-bonds, we can assume that larger structural motions occur in unliganded simulations. However, we are interested to know whether any trends exist within the different segments of the oligomer. Therefore, we examined the root means square fluctuations (RMSFs) of the last 50 ns in the 24 chains for all simulations to uncover the nature of structural motion related to each chain, to determine the role of the chemical compounds, and to identify the segments with the most flexible chains. In the unliganded simulations, chains 1 and 6 showed RMS fluctuations of roughly 5 Å, without considering the residues at N or C-ter, which naturally are flexible especially if they are disordered (Fig. 3c). The regions with the most significant motions were in chains 1 and 6. Of these, chain 1 exhibited higher values primarily in three areas: aa 40 to 46, aa 56 to 63, and aa 79 to 84 (note that that segments are not necessarily loop). We view these regions where RMS fluctuation began to increase to a maximum fluctuation of ~ 4 and ~ 4.8 Å for chain 1, as well as ~ 3 Å and 2.5 Å for chain 6 for aa 40 to 46, and aa 79 to 84. Other chains (2 and 5) exhibited lower motion compared to chains 1 and 6 in the unliganded simulation, Gin and C10 (Fig. 3c-d and Supp Fig. 2–3).

Chain fluctuation in C14 was lower than that in the unliganded simulation. Still, the trend of highly flexible regions resembled that of the unliganded simulation, with residues of aa 44 to 46, aa 56 to 63, and aa 82 to 84. Whether these flexible regions have a biological role or function as a critical protein segment has yet to be determined (Fig. 3d and Supp Fig. 2–3).

We then analyzed the protein's conformational dynamics and collective motion using Essential Dynamics (ED). We were able to calculate most of the converged conformations from 250 ns to 300 ns. The complexity of the generated data during MD simulation needed to be addressed with effective analysis to define the biological value of the conformational changes. The high-frequency localized motion of the proteins may be less crucial to the function than collective motions. We used Principal Components Analysis (PCA) to reduce the high-dimensional dataset onto the functional collective coordinates, as has been done successfully in several studies. Covariance matrixes have been generated from 250 ns to 300 ns for the Cα atoms in each chain. The diagonalization of the covariance matrix yielded the eigenvalues, then PC subspace spanned PC1 and PC2 for the unliganded, Gin, and C14 simulations for all six chains compared in each simulation (Fig. 4a-f).

A small number of PC vectors represented most of the movements of the chains. Here, we considered the first three PCs of the evaluated data. To observe the conformational dynamics of each trajectory along the eigenvectors, we projected the trajectories on the first two PCs. Chain 1 in the unliganded, Gin, and C14 simulations had more extended conformations that occupied a large phase space compared to chains 2 or 3. Chain 1 was also more flexible and may be described as more representative of the protein's function in its interactions with various factors (Fig. 4a-c). However, their conformational behaviors were different; this was especially obvious when we compared the unliganded and two liganded systems, which further can be extended to the role of the ligand. We observed more restricted motion in the essential subspace, which also correlated with lower RMS fluctuation, in chains 2 and 3. This was unsurprising because their motion was limited through the constrained sides of chains 1 and 4. In all three simulations, chains 1 and 6 mapped different areas of the principal plane, and chains 2 through 5 of the liganded systems occupied closer regions on the plane compared to those in the unliganded system (Fig. 4d-f).

We generated a heat map to show the interaction energy from 250 ns to 300 ns in the Ulig, Gin, and C14 simulations to elucidate the energetic contributions of each chains’ residue with neighboring chain and to evaluate their roles in stabilizing the protein. In each chain (chains 1–6), all residues, one by one, are considered as one group and the neighboring chains as the second group, we then calculate the L-J and Coulomb interactions between each residue and its adjacent chain to get the interaction energy (IE) which is the sum of L-J and Coulomb. The primary assessment showed that the IE among all liganded chains was higher than that of the unliganded chains (Table 1).

We divided α-Syn into four segments based on the extent to which they were ordered or where the shape turn began from the chain. Segment 1 encompassed aa 37 to aa 48, which were full loops; segment 2 encompassed aa 49 to aa 58, fully developed into a sheet conformation; segments 3 and 4, which encompassed aa 59 to aa 79 and aa 80 to aa 99, respectively, composed of loop and small sheet conformations. (Supp Fig. 4)

In chain 1, segment 1 had a lower IE value in the unliganded simulation for almost all residues apart from Val48, which strongly contributed to the binding to chain 2 and had an IE of -54.8 kJ/mol. The IE between residues of chains 2–3, 3–4, 4–5, and 5–6 for segment 1 differed between the Gin and C14 simulations, but showed a similar trend in that most regions of the heat map were reddish in color. The exceptions were four residues (37 through 40) of IE between the residues of chains 3–4 and 5–6, which interacted relatively strongly with their neighboring chains. (Table 1). The sum of the IE of all segments’ 1 residues between chains 1–2, 2–3, 3–4, 4–5, and 5–6 in the unliganded simulation were -166.1, -338.4, -269.9, -245.5, and − 347.9 kJ/mol, respectively. The weakest IE was between chains 1 and 2, and the most robust was between chains 5 and 6. Chain 1 only interacted in one direction with chain 2, has more directional freedom compared to other chains exhibited different behaviors. The sums of the IE for segment 1 between chains 1–2, 2–3, 3–4, 4–5, and 5–6 in C14 were higher in all chains than in the unliganded simulation, apart from between chains 2–3 and higher than chains in Gin apart from between chains 3–4. This is noteworthy to mention that C14 is among those compounds that interact inside the protofibril.

Two residues that strongly contributed to stability in the interactions between the six chains, were Thr44 and Glu46 with the sums of -558.2 and − 736.1 kJ/mol, respectively, across all simulations, which are the higher among all segment 1 residues in all three simulations. This might suggest that these two residues or the region that cover them can be a potential site of targeting, especially peptide-based or monoclonal antibodies against α-Syn aggregates (Table 1).

In segment 2 (residues 49 through 58), some residues had lower IE in all simulations compared to others. For the six chains, His50, Ala53, Val55, and Glu57 (apart from Unlig) had very low IE at -315.3, -173.4, -244.5, and − 332.5 kJ/mol (sum), respectively, if we exclude the contribution of Glu57 in the unliganded simulation at -62.6kJ/mol. Unsurprisingly, the sum of all the residues IE in segment 1 for the three simulations was − 4742.4, compared to -5358.3 kJ/mol in segment 2. These results indicate that the sheets in segment 2 are tied more strongly together, are among the significant points for fibril stability, and might represent the pivot points for the aggregate. In segment 3 (aa 59 through 79), small sheets and loops were the most significant constituents that makes this segment the most challenging target because of its nature and drug accessibility. The sum of the IE for all chains in the three simulations for this segment was -12011.6 kJ/mol, which is higher than the sums for segments 1 and 2 at -10100.3 kJ/mol (segments 1 and 2 compose of 22 residues compared to 21 residues in segment 3). This shows that the combination of loop and sheet can strongly affect the strength of the interactions between chains. This segment and its residues may not be a reasonable target for the design of an effective therapy. Further, in segment 3, the IE between chains (chains 1 and 2, 2 and 3, 3 and 4, 4 and 5, 5 and 6) were weaker in the unliganded simulation. The lowest IE values were observed between chains 2 and 3 at -596.22 kJ/mol, in Unlig compared to -818.32 kJ/mol, and − 917.47 kJ/mol in the Gin, and C14 simulations, respectively. (Table 1).

Segment 4 comprised residues with a dominant loop-like shape. The sum of their IE in the three systems was lower than that in segment 3. The weakest IE was observed at -414.4 kJ/mol between chains 5 and 6 in the unliganded simulation. Residues of these segments in all three simulations and across all chains had very similar and weak IE; among them, the sums of Gly84, Gly86, Ala89, and Ala91 were − 150.3, -168.1, -164.1, and − 159.8 kJ/mol, respectively, for all chains in the three simulations.

Oligomers of small size and low molecular weight play the important role of transient state in generating larger fibrils and aggregates. They are a transient and intermediate species, and reports exist of their toxicity. Our study shows that even when all chains were made from identical compositions of amino acids, the strengths of the interactions between the chains in the four defined segments differed. This result is not surprising, considering the dynamism of the proteins and the involvement of other factors, such as chemical agents, waters, and ions. However, how these dynamitists can be interpreted or fully addressed in detail to benefit the design of an effective strategy to prevent transition into full fibrils still needs to be discovered.

Applying and exposing the cells to drugs at low concentrations did not necessarily induce the destabilization of the oligomers and aggregate. We must consider the limitations of computational studies, the concentrations of chemicals, and their compositions. Our evaluation of the structural motion of chains and the H-bonds between chains shows that tested chemical compounds generate fewer conformational motions and more H-bond interactions between oligomer chains. This may not be easily applicable to all chemicals. However, many chemicals that have been tested without success might increase stability by facilitating more interactions among the chains of the protein.

We may have uncovered a trend among the more flexible residues (aa 44–46, aa 56–63, aa 82–84) in that they resembled flexible regions in the unliganded simulation. Both computational and wet lab experiments have limitations. Structural and mechanistic information on the role of the residues from each chain offers a more robust strategy for encountering aggregation and preventing it. The ED evaluation shows that the conformational dynamics of most chains in the liganded simulation are not like unliganded simulations due to the impact of the chemical compounds. Still, these diverse conformations, which map differently at the plane, do not necessarily have weaker interactions among their chains, which might explain why there are still ineffective medications against Parkinson's or other neurodegenerative disorders. This study evaluated the contribution of 62 amino acids of all chains.

In our study, we divided the core of the oligomer into four segments from the point where the loop changed to a sheet, or a sharp turn began in the conformation. Segments 1 to 4 were aa 37 through aa 48, aa 49 through aa 58, aa 59 through aa 79, and aa 80 through 99, respectively. Val48 exhibited the lowest IE in the unliganded simulation for all residues (apart from chain 1). This may have been the result of conformational freedom or compound impact. In the Gin and C14 simulations, chain 1 had the same dimensional freedom as in the unliganded simulation, but only in the unliganded simulation did it show its weakest interaction with chain 2 for Val48. The sum of the IE between neighboring chains was higher in the C14 simulation than in the unliganded one, with the exception of the IE between chains 2 and 3, and higher than in the Gin simulation apart between chains 3 and 4. Residues such as His50, Ala53, Val55, and Glu57 (apart from Unlig) for the six chains exhibited very low IE at -315.3, -173.4, -244.5, and − 332.5 kJ/mol (sum), respectively, if we exclude the contribution of Glu57 in the unliganded simulation at -70.6, which showed the substantial contribution of the residue in the unliganded simulation for segment 2. In segment 3 (aa 59–79), the sum of the IE for all chains was higher than the sums in segments 1 and 2. Small sheets and loops were significant constituents that made it the most challenging target because of its nature and drug accessibility; we can conclude that drug may not easily penetrate this region. The sum of the IE for all chains in the three simulations for this segment was − 12011.6 kJ/mol, which was higher than the sums of segments 1 and 2 at -10100.3 kJ/mol. This shows that the combination of loops and sheets can strongly affect the strength of the interactions between chains.

Many chemical compounds have been tested against PD with no success, most of these have failed to generate a robust impact on α-Syn aggregates. Some may have a partial impact but force the stabilization of protofibrils. We believe that conducting computational investigations before synthesis and wet lab testing significantly reduces the cost and time required for chemicals to reduce the density of the protein aggregates present in neurodegenerative disorders, especially PD. We conclude that candidates with extended aliphatic chains may be able to stabilize the non-bonded interactions among chains and strengthen their interactions.

Acknowledgements

The Graham-cluster, Digital Research Alliance of Canada is acknowledged as a computing resources.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Sulzer D, Alcalay RN, Garretti F, Cote L, Kanter E, Agin-Liebes J, et al. T cells from patients with Parkinson’s disease recognize α-synuclein peptides. Nature. 2017;546(7660):656-61.
Kalia LV, Lang AE. Parkinson's disease. Lancet. 2015(1474-547X (Electronic)).
Spillantini MG, Crowther Ra Fau - Jakes R, Jakes R Fau - Hasegawa M, Hasegawa M Fau - Goedert M, Goedert M. alpha-Synuclein in filamentous inclusions of Lewy bodies from Parkinson's disease and dementia with lewy bodies. Proc Natl Acad Sci U S A. 1998(0027-8424 (Print)).
Polymeropoulos MH, Lavedan C, Leroy E, Ide SE, Dehejia A, Dutra A, et al. Mutation in the α-Synuclein Gene Identified in Families with Parkinson's Disease. Science. 1997;276(5321):2045-7.
Shahmoradian SH, Lewis AJ, Genoud C, Hench J, Moors TE, Navarro PP, et al. Lewy pathology in Parkinson’s disease consists of crowded organelles and lipid membranes. Nature Neuroscience. 2019;22(7):1099-109.
Winn HR. Youmans and Winn Neurological Surgery, . 7 ed: Elsevier; 2017.
DaSilva KA, Shaw JE, McLaurin J. Amyloid-β fibrillogenesis: Structural insight and therapeutic intervention. Experimental Neurology. 2010;223(2):311-21.
Wang Z, Gao G, Duan C, Yang H. Progress of immunotherapy of anti-α-synuclein in Parkinson’s disease. Biomedicine & Pharmacotherapy. 2019;115:108843.
Choong C-J, Mochizuki H. Neuropathology of α-synuclein in Parkinson's disease. Neuropathology. 2022;42(2):93-103.
Serratos IN, Hernández-Pérez E, Campos C, Aschner M, Santamaría A. An Update on the Critical Role of α-Synuclein in Parkinson’s Disease and Other Synucleinopathies: from Tissue to Cellular and Molecular Levels. Molecular Neurobiology. 2022;59(1):620-42.
Mori A, Imai Y, Hattori N. Lipids: Key Players That Modulate α-Synuclein Toxicity and Neurodegeneration in Parkinson’s Disease. International Journal of Molecular Sciences [Internet]. 2020; 21(9).
Lashuel HA, Overk Cr Fau - Oueslati A, Oueslati A Fau - Masliah E, Masliah E. The many faces of α-synuclein: from structure and toxicity to therapeutic target. 2013(1471-0048 (Electronic)).
Lavatelli F, di Fonzo A, Palladini G, Merlini G. Systemic amyloidoses and proteomics: The state of the art. EuPA Open Proteomics. 2016;11:4-10.
Araki K, Yagi N, Aoyama K, Choong C-J, Hayakawa H, Fujimura H, et al. Parkinson’s disease is a type of amyloidosis featuring accumulation of amyloid fibrils of α-synuclein. Proceedings of the National Academy of Sciences. 2019;116(36):17963-9.
Iwai A, Masliah E, Yoshimoto M, Ge N, Flanagan L, Rohan de Silva HA, et al. The precursor protein of non-Aβ component of Alzheimer's disease amyloid is a presynaptic protein of the central nervous system. Neuron. 1995;14(2):467-75.
Chen J, Zaer S, Drori P, Zamel J, Joron K, Kalisman N, et al. The structural heterogeneity of α-synuclein is governed by several distinct subpopulations with interconversion times slower than milliseconds. Structure. 2021;29(9):1048-64.e6.
Palazzi L, Bruzzone E, Bisello G, Leri M, Stefani M, Bucciantini M, et al. Oleuropein aglycone stabilizes the monomeric α-synuclein and favours the growth of non-toxic aggregates. Scientific Reports. 2018;8(1):8337.
Ellis JM, Fell MJ. Current approaches to the treatment of Parkinson’s Disease. Bioorganic & Medicinal Chemistry Letters. 2017;27(18):4247-55.
Javed H, Nagoor Meeran MF, Azimullah S, Adem A, Sadek B, Ojha SK. Plant Extracts and Phytochemicals Targeting α-Synuclein Aggregation in Parkinson's Disease Models. Frontiers in Pharmacology. 2019;9.
Uversky VN. Neuropathology, biochemistry, and biophysics of alpha-synuclein aggregation. J Neurochem. 2007(0022-3042 (Print)).
Sarrafchi A Fau - Bahmani M, Bahmani M Fau - Shirzad H, Shirzad H Fau - Rafieian-Kopaei M, Rafieian-Kopaei M. Oxidative stress and Parkinson's disease: New hopes in treatment with herbal antioxidants. Curr Pharm Des 2016Curr Pharm Des.(1873-4286 (Electronic)).
Wang ZY, Liu JY, Yang CB, Malampati S, Huang YY, Li MX, et al. Neuroprotective Natural Products for the Treatment of Parkinson's Disease by Targeting the Autophagy-Lysosome Pathway: A Systematic Review. Phytother Res. 2017(1099-1573 (Electronic)).
Ghiglieri V, Calabrese V, Calabresi P. Alpha-Synuclein: From Early Synaptic Dysfunction to Neurodegeneration. Front Neurol. 2018(1664-2295 (Print)).
Broersen K, Ruiperez V, Davletov B. Structural and Aggregation Properties of Alpha-Synuclein Linked to Phospholipase A2 Action. Protein Pept Lett. 2018(1875-5305 (Electronic)).
Goux WJ, Kopplin L, Nguyen AD, Leak K, Rutkofsky M, Shanmuganandam VD, et al. The Formation of Straight and Twisted Filaments from Short Tau Peptides *. Journal of Biological Chemistry. 2004;279(26):26868-75.
Pieri L, Madiona K, Melki R. Structural and functional properties of prefibrillar α-synuclein oligomers. Scientific Reports. 2016;6(1):24526.
L P, K M, L B, R M. - Fibrillar α-synuclein and huntingtin exon 1 assemblies are toxic to the cells. D - 0370626. 2012(- 1542-0086 (Electronic)):- 2894-905.
L P, K M, R M. - Structural and functional properties of prefibrillar α-synuclein oligomers. D - 101563288. 2016(- 2045-2322 (Electronic)):- 24526.
N C, SI C, E D, AY A, AY C, A O, et al. - Direct observation of the interconversion of normal and toxic forms of. D - 0413066. 2012(- 1097-4172 (Electronic)):- 1048-59.
Zhu M, Rajamani S, Kaylor J, Han S, Zhou F, Fink AL. The Flavonoid Baicalein Inhibits Fibrillation of α-Synuclein and Disaggregates Existing Fibrils. 2004;279(26):26857.
MJ V, Lansbury PT J. - Zeroing in on the pathogenic form of alpha-synuclein and its mechanism of. D - 0370623. 2003 (- 0006-2960 (Print)):- 7871-8.
Pujols J, Peña-Díaz S, Lázaro DF, Peccati F, Pinheiro F, González D, et al. Small molecule inhibits α-synuclein aggregation, disrupts amyloid fibrils, and prevents degeneration of dopaminergic neurons. Proceedings of the National Academy of Sciences. 2018;115(41):10481-6.
Angelopoulou E, Paudel YN, Papageorgiou SG, Piperi C. Elucidating the Beneficial Effects of Ginger (Zingiber officinale Roscoe) in Parkinson’s Disease. ACS Pharmacology & Translational Science. 2022;5(10):838-48.
Hsu K-C, Chen Y-F, Lin S-R, Yang J-M. iGEMDOCK: a graphical environment of enhancing GEMDOCK using pharmacological interactions and post-screening analysis. BMC Bioinformatics. 2011;12(1):S33.
Spoel LAHvd. GROMACS 2021.5. 2021
Abraham MJ, Murtola T, Schulz R, Páll S, Smith JC, Hess B, et al. GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX. 2015;1-2:19-25.
Huang J, Rauscher S, Nawrocki G, Ran T, Feig M, de Groot BL, et al. CHARMM36m: an improved force field for folded and intrinsically disordered proteins. Nature Methods. 2017;14(1):71-3.
Best RB, Zhu X, Shim J, Lopes PEM, Mittal J, Feig M, et al. Optimization of the Additive CHARMM All-Atom Protein Force Field Targeting Improved Sampling of the Backbone ϕ, ψ and Side-Chain χ1 and χ2 Dihedral Angles. Journal of Chemical Theory and Computation. 2012;8(9):3257-73.
Koziara KB, Stroet M, Malde AK, Mark AE. Testing and validation of the Automated Topology Builder (ATB) version 2.0: prediction of hydration free enthalpies. Journal of Computer-Aided Molecular Design. 2014;28(3):221-33.
Darden T, York D, Pedersen L. Particle mesh Ewald: An N⋅log(N) method for Ewald sums in large systems. The Journal of Chemical Physics. 1993;98(12):10089-92.
Bussi G, Donadio D, Parrinello M. Canonical sampling through velocity rescaling. The Journal of Chemical Physics. 2007;126(1):014101.
Berendsen HJC, Postma JPM, van Gunsteren WF, DiNola A, Haak JR. Molecular dynamics with coupling to an external bath. The Journal of Chemical Physics. 1984;81(8):3684-90.
Parrinello M, Rahman A. Polymorphic transitions in single crystals: A new molecular dynamics method. Journal of Applied Physics. 1981;52(12):7182-90.
Kumari R, Kumar R, Lynn A. g_mmpbsa—A GROMACS Tool for High-Throughput MM-PBSA Calculations. Journal of Chemical Information and Modeling. 2014;54(7):1951-62.
Baker NA, Sept D, Joseph S, Holst MJ, McCammon JA. Electrostatics of nanosystems: Application to microtubules and the ribosome. Proceedings of the National Academy of Sciences. 2001;98(18):10037.
Lipinski CA. Lead- and drug-like compounds: the rule-of-five revolution. Drug Discovery Today: Technologies. 2004;1(4):337-41.
Jayaram B, Singh T, Mukherjee G, Mathur A, Shekhar S, Shekhar V. Sanjeevini: a freely accessible web-server for target directed lead molecule discovery. BMC Bioinformatics. 2012;13(17):S7.

Table 1 is available in the Supplementary Files section.

No competing interests reported.

Download PDF

Journal Publication

published 03 Nov, 2023

Read the published version in Scientific Reports →

Editorial decision: Major revision
12 Sep, 2023
Reviews received at journal
30 Jul, 2023
Reviewers agreed at journal
16 Jul, 2023
Reviews received at journal
11 Jul, 2023
Reviewers agreed at journal
11 Jul, 2023
Reviewers invited by journal
11 Jul, 2023
Editor assigned by journal
11 Jul, 2023
Editor invited by journal
11 Jul, 2023
Submission checks completed at journal
11 Jul, 2023
First submitted to journal
03 Jul, 2023

You are reading this latest preprint version

Conformational Dynamics of α-Synuclein: A Study of its Intramolecular Forces in the Presence of Selected Compounds

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials and Methods

3.1. Docking Simulation

3.2. Molecular Dynamics Simulation

3.3. Free Energy Evaluation

Result and Discussion

Discussion

Declarations

References

Table

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1