Effect of Familial Mutations on the Interconversion of Alpha Helix to

Subscriber access provided by United Arab Emirates University | Libraries Deanship

Article

Effect of Familial Mutations on the Interconversion of Alpha Helix to Beta Sheet Structures in Amyloid Forming Peptide : Insights from Umbrella Sampling Simulations Sathish Kumar Mudedla, N. Arul Murugan, and Hans Ågren ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.8b00425 • Publication Date (Web): 26 Dec 2018 Downloaded from http://pubs.acs.org on January 1, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

Effect of Familial Mutations on the Interconversion of Alpha Helix to Beta Sheet Structures in an Amyloid Forming Peptide: Insight from Umbrella Sampling Simulations Sathish Kumar Mudedla,a,* N. Arul Murugana,* and Hans Agrena,b a

Division of Theoretical Chemistry and Biology, School of Biotechnology, AlbaNova University Center, Royal Institute of Technology (KTH), S-106 91 Stockholm, Sweden bCollege of Chemistry and Chemical Engineering, Henan University, Kaifeng, Henan 475004P. R. China * To whom correspondence should be addressed. E–mail: [email protected], [email protected] Abstract Understanding the initial events in aggregation of amyloid beta monomers to form beta sheet rich fibrils is useful for the development of therapeutics for Alzheimer’s disease. In this context, the changes in energetics involved in the aggregation of helical amyloid beta monomers into beta sheet rich dimers have been investigated using umbrella sampling simulations and density functional theory calculations. The results from umbrella sampling simulations for the free energy profile for the interconversion are in close agreement with density functional theory calculations. The results reveal that helical peptides converted to beta sheet structures through coil-like conformations as intermediates which are mostly stabilized by intramolecular hydrogen bonds. The stabilization of intermediate structures could be a possible way to inhibit the fibril formation. Mutations substantially decrease the height of the energy barrier for interconversion from alpha-helix to beta-sheet structure when compared to wild type something that is attributed to an increase in number of intramolecular hydrogen bonds between backbone atoms in the coil structures which correspond to a maximum value on the free energy surface. The reduction of the energy barrier leads to an enhancement of the rate of aggregation of amyloid beta monomers upon various familial mutations which concords with previous experimental reports.

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 27

Keywords: Amyloid forming peptide, mutation effect on rate of aggregation, Alzheimer’s disease, Free energy for alpha-helix to beta interconversion, umbrella sampling simulations, Familial mutations

wild type slow

Conversion

mutation α-helix faster

β-sheet

Introduction Alzheimer’s disease (AD) is a commonly seen neurodegenerative disorder in old aged people (older than 65 years) and less commonly in people with age below 65 years.1,

2

It

causes memory loss, agitation, withdrawal and inability to respond to stimuli.3 It is commonly attributed to the aggregation of amyloid beta peptides and neurofibrillary tangles in the brain making the deposition of amyloid fibrils one of the pathological hallmarks of AD.4 The amyloid beta peptide is produced from amyloid precursor protein by sequential cleavage by enzymes β and γ-secretases.5 The misfolded amyloid beta peptide monomers aggregate to beta sheet rich dimers, trimers and oligomers with subsequent formation of larger plaques.6 The busting of aggregated amyloid fibrils, inhibition of misfolding and


Page 3 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


aggregation of amyloid peptides constitute promising actions to treat AD.7-12 The intermediate sized oligomers which are often amorphous in shape are the most neurotoxic forms of amyloid beta peptides as they can disrupt cell membranes.13, 14 The high resolution molecular structures for amyloid beta oligomers are not yet known due to their transient nature. Recently, the structural models for dimers of amyloid beta have been developed using molecular dynamics simulations.15 NMR experiments have shown that the amyloid beta monomers attain random coil conformations in aqueous solutions,16-18 while in non-polar environments the same peptides have significant helical content.19-21 Amyloid beta monomers may exist in random coil or helix conformation to initiate the oligomerization process. The early events of oligomerization process are still unexplored due to their intrinsic disorder and high propensity to further aggregation. Understanding the initial events of oligomerization from monomers could thus contribute to the initial development of therapeutics of AD. The pathogenic mutations in amyloid precursor proteins occur in the amyloid forming segments of the protein.22 The most common pathogenic familial mutations are Flemish, Iowa, Arctic, Italian and Dutch.23-27 Flemish and Iowa are caused by the replacement of 21 amino acid (alanine (A)) to glutamic acid (G) and 23 residue aspartic acid (D) with asparagine (N), respectively. Arctic, Italian and Dutch are produced due to glutamic acid (E) at 22 position switched with glycine (G), lysine (K) and glutamine (Q), respectively. These mutations cause early onset of AD and changes the physicochemical properties of amyloid beta peptides. The neurotoxicity of mutated fibrils is higher than that of wild type.28 Based on the THT staining experiments, is has been found that the rate of aggregation of amyloid beta peptides also increases due to mutation, except the Flemish case in which the process is slower than for the wild type.29-32 Molecular dynamics simulations performed by Panchal et al. have indicated an instability of the fibril after mutations when compared to wild type.33 Pande et al. have characterized the



consequences of monomer structures on mutations and shown that increased helix propensity for Italian and Dutch whereas the same decreased in the case of Iowa and Arctic.34 Another previous report has unravelled the effect of familial mutations on the structure of nucleation site for folding using amyloid beta peptide with residues in the range 21-30.35 The mutations involving glutamic acid do not change the structure of site where nucleation for folding occurs but it may affect the long range interaction between the monomeric peptides and therefor the aggregation rate. In the case of Iowa mutation leads to the turn conformation for folding nucleation site that may influence further aggregation kinetics. In another molecular dynamics simulation study, it was shown that the variations in interactions between the central hydrophobic part (17-21) and the bend region (22-28) after mutation with glutamine at 22 position is the reason for the change in kinetics when compared to wild type.36 The changes in kinetics for Taiwan mutation (D7H) have been explained using the reduction of beta sheet content and salt bridge interactions.37 However, the mechanism of aggregation of wild type amyloid forming peptides and the effect of mutations on the aggregation rate is still not well established. The reasons for the enhanced rate of the fibrillation process (exception being the Flemish mutation) in the case of familial mutations are still elusive. Therefore, in this study, we have studied the free energies involved in the conversion of alpha helix to beta sheet structures in the dimers of an amyloid forming sequence KLVFFGEDV through umbrella sampling simulations.

Also the peptide sequences associated with familial

mutations namely Flemish, Iowa, Arctic, Italian and Dutch are considered. The energetics from force-field methods for the interconversion process, have been also verified using density functional theory calculations. Results and Discussion The aim of the study is investigate the driving force for interconversion of α-helical peptides to β-sheet conformation in the amyloid forming sequence and to estimate the mutation effect


Page 4 of 27

Page 5 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


on this interconversion. The secondary structures of the peptides depends on the psi and phi backbone angles. The psi angles for the helix and beta sheet conformations are approximately -40˚ and 120˚ whereas the range of phi angles is similar for both cases. A previous study has shown that the average psi can be used as a reaction coordinate for the conversion of helix to beta sheet conformation using umbrella sampling and metadynamics simulations.45 In this study, we make use of the same concept to bring in the changes in the secondary structure for a small amyloid forming peptide with sequence KLVFFGEDV and its single point mutants corresponding to the cases of different familial mutations. Each peptide in the case of dimer has 8 psi angles - the average value is set to 120˚ for the starting structure which is in beta sheet form. First, we have performed restrained simulations with an average psi of 120˚ for 10 ns. The structure at the end of this simulation was used to obtain conformations related to the average psi of 130˚. A similar procedure was applied to sample the conformational space between -180˚ and 180˚ in a stepwise fashion by varying the average psi angle by -10˚ or 10˚, respectively. We have simulated 37 configurations with a total simulation time of 370 ns. We varied the average psi angle from 120˚ to 180˚ (and 120˚ to -180˚) with an increase (or decrease) of 10˚ in order to sample the different conformations of amyloid beta peptides. For each average psi angle, a restrained simulation was performed for 10 ns using harmonic potentials. During these simulations, we observed that the dimer with β-sheet conformation converts to α-helix structure as we reach an average psi angle around -40. This interconversion is not only seen for the wild type but also in all mutants studied here. The peptides attain different secondary structures such as coil, β-sheet, bend, turn, α-helix and 310helix at various average psi angles. The percentage of different secondary structural elements in the wild type and mutant variants at different psi angles are given in Table 1-6 (of the supporting information). The peptides dominantly exist in α-helical conformation when the average psi angle is in the range -20˚ to -70˚ and it is most prominent for the angles -40˚ and -



50˚. The helical content in wild type, A21G, D23N, E22G, E22K and E22Q is 73, 71, 72, 73, 73 and 71%, respectively. As can be seen, the difference in the helical content of the mutated peptides when compared to wild type is marginal and the mutations do not affect the secondary structure contents of the helical conformation. It is observed that the beta sheet content varies significantly for the mutants with respect to wild type and each mutated system adopts beta-sheet like structure at different psi angle. The beta sheet conformations were adopted by the wild type and mutants for the average psi angle range of 160˚ to 90˚. The predominant beta sheet contents are 66% (corresponding to average psi angle, 100˚), 48% (90˚), 52% (110˚), 62% (110), 67% ( 100˚ ), 64% ( 110˚ ) for wild type, A21G, D23N, E22G, E22K and E22Q, respectively. As can be seen, the mutations reduce the beta sheet content in the dimeric form of amyloid beta than that of wild type except for E22K. The lowest beta sheet content is noted in the case of A21G. The secondary structure for the intermediate conformer appearing between α-helix and beta sheet is mostly in coil form. It is clear that the helical amyloid peptides transform to disordered coil form then they form beta sheet rich structures. Mutation does not change the mechanism involved in the conversion of α-helix to beta sheet aggregates and invariably coil-like structures appeared as intermediates also in the mutated cases. To understand the relative energetics of the different secondary structures such as helix, coil and beta-sheet appearing along the interconversion pathway, we have calculated the free energy profile as a function of average psi angle for all systems (wild type and mutants). The free energies are based on the WHAM analysis carried out using the umbrella sampling simulations. The calculated free energies for the mutated systems are compared with wild type, see Figure 1. It can be clearly noted that each free energy surface has minima around at -40˚ and 140˚ which corresponds to α-helix and beta sheet conformations, respectively. The helix has lower energy when compared to beta sheet conformation which suggests that the


Page 6 of 27

Page 7 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


former secondary structure is the most stable one for the dimeric form. Probably, for the larger oligomers this trend may change making the beta-sheet structure the most stable secondary structure. We have analysed the reason why the two secondary structures, namely helix and beta-sheet, are the most stable. We found that helical structures at -40˚ are stabilized by the intermolecular hydrogen bonds whereas the beta sheet conformations are stable due to intermolecular hydrogen bonds. We found an intermediate minimum in between the helix and beta sheet structures around at 50˚. The two minima (at 50˚ and 140˚) are separated by one maximum which corresponds to coil structure and is located around 80˚. The aggregation of amyloid peptides thus involves conversion of helix to coil structure and the latter further forms beta sheet conformations. The energy barrier for the transition of helix to coil is 1.42, 0.66, 1.39, 1.77, 1.96 and 1.44 kcal/mol for wild type, A21G, D23N, E22G, E22K and E22Q, respectively. The energy barrier reduces on mutation except in the case of E22G, E22K and E22Q. Upon further analysis, we found that the height of maximum between coil and beta sheet structure is lowered in the case of mutant dimers when compared wild type. This shows that the mutation reduces the energy barrier for the conversion of coil to beta sheet conformation - the energy barriers are 0.85, 1.07, 0.72, 0.89, 0.79 and 0.64 kcal/mol for wild type, A21G, D23N, E22G, E22K and E22Q, respectively. The total energy barrier involved in the conversion of helix to beta sheet conformation is 2.27, 1.73, 2.11, 2.66, 2.75 and 2.08 kcal/mol for wild type, A21G, D23N, E22G, E22K and E22Q, respectively. The calculated energy barriers for helix to beta-sheet conformation are thus significantly lower and one may expect such interconversion to be feasible from routine molecular dynamics calculation without any enhanced sampling method. However, in order to achieve the secondary structure change from alpha-helix to beta-sheet all these psi values should be set to a specific range (approximately -40 to 120) which makes the conversion of helix to beta sheet a rare event for the system to explore. If we assume 1 ps for a



conformational jump and if there are 16 peptide bonds one can expect approximately (316)*0.001 ns (3 corresponds to number of minimum energy conformers namely, gauche, gauche- and trans and here we have not included the conformational space due to phi angles) as the time scale for a system to reach a specific conformation. Even though the energy barrier as obtained from umbrella sampling simulations is not too high in order to arrive at the specific conformation, the system should make a move such a way that all the psi angles adopt a specific value which makes this event not easily occurring in constrained free MD simulations. The order of energy barrier follows as E22K > E22G > Wild type > D23N > E22Q > A21G. However, experimentally it is reported that only in the case of A21G the rate of fibrillation is lowered when compared to wild type and for all other mutations the rate of aggregation is increased when compared to wild type. The free energy calculations as a function of average psi angle could correctly reproduce the trend in the fibrillation rate for the mutants D23N and E22Q. However, the results for the fibrillation rate for the cases E22G, E22K and A21G are not well reproduced and it may be attributed to force-field which is unable to capture effect of the mutations on the aggregation rate. Further to understand the energetics of the structures at different psi angles and to reproduce the free energy surface, we have calculated the quantum mechanical energies using density functional theory. The calculated total energies were obtained by averaging over 10 structures which are collected from the molecular dynamics trajectories corresponding to different selected psi angles. The calculated relative energies are shown in Figure 2. The energy values show that the two minima correspond to helix and beta sheet structures and maximum correspond to coil structure. It is corroborated with the analysis of free energy surface from the WHAM analysis method. However, we found the shifts in position of minimum and maximum on the energy surface when compared to the free energy surface as obtained from WHAM analysis something that may due to inclusion of electronic effects. It can be seen that


Page 8 of 27

Page 9 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the lowest minimum exists at -40˚( helix) for all the cases. The minimum corresponds to beta sheet conformations at 100˚ and 110˚ which attributes to high beta sheet content. This result is in correlation with the secondary structural details from umbrella sampling simulations. The energy surface has another minimum located between 50˚ and 70˚. And also the maximum in free energy value exists in the range from 20˚ and 40˚ and at 80˚. The helical, coil and beta sheet conformations of all dimeric forms are shown in Figure 3. The coil structures (between 50˚ and 70˚) and beta sheet forms are separated with a maximum at 80˚. This shows that helical peptides transform to beta sheet aggregates via intermediate coil structures. Earlier it has been shown from metadynamics simulations that amyloid forming Prion proteins also aggregate into beta sheet conformations through the formation of a random coil structure as an intermediate.45 It is worth noting that wild type and mutated amyloid peptides follow the similar mechanism in order to form amyloid aggregates. Interestingly, we notice that the maximum energies are lower in magnitude for mutated dimers when compared to wild type (except in the case of E22K). The reduction in energy barrier eventually leads to a faster formation of the beta sheet aggregates which could explain the observed mutation induced increased rate of aggregation in this peptide. Moreover, in contrast to force-field methods, the beta sheet structure of E22K is more stable than wild type and the same is found for the other mutated peptides except for A21G. In the case of A21G, the energies of the beta sheet structures are thus high when compared to the wild type. This may lead to a slower rate of aggregation process than that of the wild type. The transition from helix to coil involves the following energy barriers for wild type, A21G, D23N, E22G, E22K and E22Q: 101.72, 75.45, 87.74, 50.12, 106.17 and 67.2 kcal/mol. Coil to beta sheet conversion are associated with the energy barriers 4.13, 16.76, 10.01, 12.37, 26.29, and 33.95 kcal/mol in the case of wild type, A21G, D23N, E22G, E22K and E22Q, respectively. The



total energy barriers for the conversion of helix to beta sheet structures follow the trend as E22K > wild type > E22Q > D23N > A21G > E22G. The large difference in free energies computed using force-field method and total energies computed using density functional theory (DFT) should be attributed to the different components to total energy in these methods. In addition, in the latter approach, the entropic contributions are not included. The barriers as obtained from DFT should have been lowered to some extend if we have performed the geometry optimization. Due to computational demand, the structures as obtained from force-field umbrella sampling simulations were used for DFT level calculations and single point energy calculations were performed. In order to see the effect of entropic contributions, we have performed normal mode calculations for wild type and mutants. The calculated formation entropies (all three translational, rotational and vibrational contributions) for the dimer of helix and beta sheet for all systems are given in Table S7 of the supporting information. We find that, as expected, there are significant differences in entropy for α-helix and β-sheet structures for the wild type and mutants. The role of translational and rotational entropies in the relative stabilization of different secondary structures is less significant. However, the vibrational contributions tend to stabilize the alpha-helix structure than the beta-sheet structure. In the umbrella sampling simulations we get free energies which are computed from the potential mean force and which implicitly accounts for entropic contributions. This analysis confirms again that the huge discrepancy in energies as we get from force-field and DFT methods is due to difference of methodology and since the entropic contributions are not accounted for in the latter case. The high stability of beta sheet structures and lowering of energy barriers in the case of D23N, E22G, E22K and E22Q could result in a fixing of the formation of amyloid beta aggregates. This may influence the total kinetics of the formation of aggregates from


Page 10 of 27

Page 11 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


amyloid beta peptides. This analysis is in agreement with previous experimental reports29-32 and the above discussed results from umbrella sampling simulations. The stability of secondary structures is attributed to the hydrogen bonds in the backbone of peptides. We have calculated the hydrogen bonds present in the wild type and mutated dimers of amyloid beta using the trajectories from the umbrella sampling simulations. The calculated number of hydrogen bonds in the dimer at each psi angle is shown in Figure 4. The minima located on the energy surface from Figure 2 has a larger number of hydrogen bonds when compared to others. The structures corresponding to minima at average psi values of 100˚, 90˚, 100˚, 110˚, 100˚ and 110˚ for wild type, A21G, D23N, E22G, E22K and E22Q have a larger number of hydrogen bonds than the other secondary structures and are in good agreement with secondary structural details and DFT energies. The helix which is a minimum at -40˚ has a larger number of hydrogen bonds than that of beta structures. This is the reason why the α-helical structure corresponds to global minimum in the free energy surface as obtained from WHAM analysis. Further, the plot shows a larger number of hydrogen bonds for the structures having an average psi angle in the range 40˚ and 60˚. It is worth noticing that the plot of DFT energies as a function of average psi angle displays a local minimum for these range of angles. A more careful analysis on the nature of hydrogen bonds suggests that these structures are stabilized by the intramolecular hydrogen bonds rather than the intermolecular hydrogen bonds. At the same time, the structures corresponding to maximum energy have a smaller number of hydrogen bonds. The maxima were noted for wild type, A21G, D23N, E22G, E22K and E22Q mutants at 40˚, 40˚, 10˚, 20˚, 20˚ and 40˚, respectively. The larger number of hydrogen bonds can stabilize the coil structures and thus may reduce the height of the energy barrier. We notice that the number of hydrogen bonds in the structures corresponding to maximum (from 20˚ to 40˚) are 9, 10, 5, 4, 3 and 7 for wild type, A21G, D23N, E22G, E22K and E22Q, respectively.

Furthermore, we calculated inter and



intra molecular hydrogen bonds for all systems at each psi angle, see Figure 5. It can be seen that inter molecular hydrogen bonds exist only in the case of beta sheet conformations and that the remaining all structures stabilized through intra-molecular hydrogen bonds between back bone atoms. The results show that the helical amyloid beta peptide transition to coil structures, which are stabilized by intramolecular hydrogen bonds, subsequently aggregate to beta sheet conformations and these are further stabilized via inter-molecular hydrogen bonds.

Since the system size is reasonably larger, we also wanted to estimate the energy profile for various configurations picked up along the reaction coordinate by using semi-empirical method, PM7. This method is relatively very fast when compared to DFT approach and so even peptides with larger length scale can as well be studied. The calculations were carried out for 25 configurations from trajectories corresponding to different average psi angles. The energy profile along the reaction coordinate for wild type and mutants are shown in Figure 6. The results are quite impressive that now all the mutants have lower energy barrier for the interconversion process from alpha-helix to beta-sheet structures and it can be suggested that semi empirical level of theory performs much better than the force-field and DFT approaches. The order of energy barrier follows as wild type>A21G>E22G > E22Q > E22K=D23N. Even though the barriers are overestimated when compared to both force-field and DFT methods, the order of energy barriers for wild type and mutants has been well reproduced. The only except being the A21G which is supposed to show higher barrier than wild type and so reduced fibrillation rate. We speculate the energy barriers should have been lowered significantly if the configurations from trajectories were optimized at PM7 level of theory.


Page 12 of 27

Page 13 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Conclusions The aggregation mechanism of small amyloid forming peptides and the effect of mutation on the kinetics of aggregation have been explored by using umbrella sampling simulations accompanied with density functional theory calculations. The pathway for the interconversion for alpha-helical to beta sheet structures in the dimeric form has been explored by varying the average psi angle which is proposed to be a suitable reaction coordinate to achieve such interconversion. The free energy surface reveals that the amyloid beta peptides transforms from helical conformation to beta sheet form via coiled structures as intermediates. The intermediate structures are stabilized with the help of intramolecular hydrogen bonds. Furthermore, familial mutations of amyloid beta do not influence the mechanism of conversion of helix to beta sheet conformation, i.e in all the cases the coil structure appears as intermediate. However, mutation lowers the height of the energy barrier when compared to the wild type which also explains the experimentally reported increase in fibrillization rate for most cases of familial mutations. We believe that this insight can give a piece to the puzzle how to inhibit aggregation through future purposeful drug design. Overall, all the methods employed display at least two minimum corresponding to alphahelix and beta-sheet like secondary structures and energy barriers vary quite significantly between different methods. The force-field method displays the lowest value for energy barrier while the semi-empirical PM7 method displays largest value. However, in terms of explaining the mutation dependent aggregation rate, the semi-empirical theory outperforms among all three methods.



Computational Details The coordinates of amyloid beta dimer has been extracted from the NMR structure of protofibril (pdb id: 2BEG).38 We have considered the sequence (16-24) which plays important role to the formation of beta sheet structures. The terminals of the extracted peptides were ended with acetyl (ACE) and N-methyl (NME) groups. We have also modelled peptides with the Flemish, Iowa, Dutch, Italian and Arctic mutations. The sequences of the considered peptides are Wild type: KLVFFGEDV, Flemish: KLVFFAEDV, Iowa: KLVFFGENV, Dutch: KLVFFGGDV, Italian: KLVFFGKDV, and Arctic: KLVFFGQDV. The mutations Flemish, Iowa, Dutch, Italian, and Arctic are represented as A21G, D23N, E22G, E22K and E22Q in the remaining text unless otherwise noted. The dimers of these modelled systems were used to study the influence of mutation on the aggregation process of amyloid beta peptide. The modelled structures solvated in a cubic box using TIP3P water model and charge is neutralized by adding Na+/Cl- ions depending upon the mutants. These systems were subjected to energy minimization using steepest decent method subsequently equilibrated for 1 nanosecond (ns) at 293 K temperature and 1 bar pressure. Velocity rescaling and Parrinello-Rahman algorithms were used to control temperature and pressure in NPT ensemble.39-41 All these simulations were performed using GROMACS package.42-44 The simulated systems have used as starting structures for umbrella sampling simulations. Each configuration was restrained using a harmonic spring constant of 150 kJ.mol-1.rad-2. All the umbrella sampling simulations were carried out using GROMACS coupled with PLUMED-2.0 software.46 The free energy profiles for the configurations used in the umbrella sampling simulations were computed with the help of weighted histogram analysis method using 18 bins.47


Page 14 of 27

Page 15 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Further, to compare the energetics obtained from weighted histogram analysis, we have calculated the total energies of structures obtained from umbrella sampling simulations using single point calculations. We have collected the structures for each 500 picosecond from the trajectories of selected psi angles. The density functional theory (DFT) was used to calculate the energies for 10 structures in the case of each psi angle. All the energies were calculated using dispersion corrected functional at B3LYP-D/6-31G* level of theory with the help of Gaussian 16 software.48 DFT calculations were carried out in water as solvent using universal solvent model which is developed by Truhlar.49 Author Contributions: The research project was designed by N. A. M and H. A. The calculations and analysis were carried out by S. K. M. The manuscript was written by S. K. M and N. A. M. H. A and N. A. M contributed to improve the quality of writing. All authors participated in the discussion of the results. Acknowledgements The authors acknowledge support from the Swedish Foundation for Strategic Research (SSF) through the project “New imaging biomarkers in early diagnosis and treatment of Alzheimer’s disease” and the support from SLL through the project “Biomolecular pro-filing for early diagnosis of Alzheimer’s disease”. This work was supported by the grants from the Swedish Infrastructure Committee (SNIC) for the projects “Multiphysics Modeling of Molecular Materials” (SNIC2017-12-49) and “In-silico Diagnostic Probes Design” (SNIC2018-3-3).



References 1. Livingston, G.; Sommerlad, A.; Orgeta, V.; Costafreda, S. G.; Huntley, J.; Ames, D.; Ballard, C.; Banerjee, B.; Burns, A.; CohenMansfield, J.; Cooper, C.; Fox, N.; Gitlin, L. N.; Howard, R.; Kales, H. C.; Larson, E. B.; Ritchie, K.; Rockwood, K.; Sampson, E. L.; Samus, Q.; Schneider, L. S.; Selbæk, G.; Teri, L.; Mukadam, N.; et al. Dementia Prevention, Intervention, and Care. Lancet. 2017, 17, 31363-31366. 2. Fratiglioni, L.; De Ronchi, D., Aguero-Torres, H. Worldwide Prevalence and Incidence of Dementia. Drugs Aging 1999, 15, 365−375. 3. Bird T.D. Genetic Factors in Alzheimer's Disease. N. Engl. J. Med. 2005, 352, 862– 864. 4. Braak, H.; and Braak, E. Neuropathological Stageing of Alzheimer-Related Changes. Acta. Acta Neuropathol. 1991, 82, 239−259. 5. Haass, C.; Selkoe, D. J. Soluble Protein Oligomers in Neurodegeneration: Lessons from the Alzheimer's Amyloid Beta-Peptide. Nat. ReV. Mol. Cell Biol. 2007, 8, 101112. 6. Hardy, J.; Selkoe, D. J. The amyloid Hypothesis of Alzheimer's Disease: Progress and Problems on the Road to Therapeutics. Science 2002, 297, 353-356. 7. Necula, M.; Kayed, R.; Milton, S.; Glabe, C. G. Small Molecule Inhibitors of Aggregation Indicate That Amyloid β Oligomerization and Fibrillization Pathways Are Independent and Distinct. J. Biol. Chem. 2007, 282, 10311−10324. 8. Walsh, D. M.; Townsend, M.; Podlisny, M. B.; Shankar, G. M.; Fadeeva, J. V.; Agnaf, O. E.; Hartley, D. M.; Selkoe, D. J. Certain Inhibitors of Synthetic Amyloid βPeptide (Aβ) Fibrillogenesis Block Oligomerization of Natural Aβ and Thereby Rescue Long-Term Potentiation. J. Neurosci. 2005, 25, 2455−2462. 9. Wang, Z.; Chang, L.; Klein, W. L.; Thatcher, G. R. J.; Venton, D. L. Per-6substituted-per-6-deoxy β-cyclodextrins Inhibit the Formation of β-Amyloid Peptide Derived Soluble Oligomers. J. Med. Chem. 2004, 47, 3329−3333. 10. Fradinger, E. A.; Monien, B. H.; Urbanc, B.; Lomakin, A.; Tan, M.; Li, H.; Spring, S. M.; Condron, M. M.; Cruz, L.; Xie, C.-W.; et al. C-terminal Peptides Coassemble Into Aβ42 Oligomers and Protect Neurons Against Aβ42-induced Neurotoxicity. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 14175−14180. 11. Yang, F.; Lim, G. P.; Begum, A. N.; Ubeda, O. J.; Simmons, M. R.; Ambegaokar, S. S.; Chen, P. P.; Kayed, R.; Glabe, C. G.; Frautschy, S. A.; et al. Curcumin Inhibits Formation of Amyloid β Oligomers and Fibrils, Binds Plaques, and Reduces Amyloid in Vivo. J. Biol. Chem. 2005, 280, 5892−5901. 12. Feng, B. Y.; Toyama, B. H.; Wille, H.; Colby, D. W.; Collins, S. R.; May, B. C. H.; Prusiner, S. B.; Weissman, J.; Shoichet, B. K. Smallmolecule Aggregates Inhibit Amyloid Polymerization. Nat. Chem. Biol. 2008, 4, 197−199. 13. Sultana, R.; Butterfield, D. A. Mol Biosyst. 2008, 4, 36‐41. 14. Kang, J.; Lemaire, H. G.; Unterbeck, A.; et al., Nature 1987, 325, 733‐736.


Page 16 of 27

Page 17 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


15. Blinov, N.; Khorvash, M.; Wishart, S. D.; Cashman, N. R.; Kovalenko, A. Initial Structural Models of the Aβ42 Dimer from Replica Exchange Molecular Dynamics Simulations. ACS Omega 2017, 2, 7621-7636. 16. Roche, J.; Shen, Y.; Lee, J. H.; Ying, J.; Bax, A. Monomeric Aβ1-40 and Aβ1-42 Peptides in Solution Adopt Very Similar Ramachandran Map Distributions That Closely Resemble Random Coil. Biochemistry 2016, 55, 762−775. 17. Yamaguchi, T.; Matsuzaki, K.; Hoshino, M. Transient formation of intermediate conformational states of amyloid-β peptide revealed by heteronuclear magnetic resonance spectroscopy. FEBS Lett. 2011, 585, 1097−1102. 18. Hou, L.; Shao, H.; Zhang, Y.; Li, H.; Menon, N. K.; Neuhaus, E. B.; Brewer, J. M.; Byeon, I.-J. L.; Ray, D. G.; Vitek, M. P.; et al. Solution NMR Studies of the Aβ(1-40) and Aβ(1-42) Peptides Establish that the Met35 Oxidation State Affects the Mechanism of Amyloid Formation. J. Am. Chem. Soc. 2004, 126, 1992−2005. 19. Tomaselli, S.; Esposito, V.; Vangone, P.; van Nuland, N. A. J.; Bonvin, A. M. J. J.; Guerrini, R.; Tancredi, T.; Temussi, P. A.; Picone, D. The α-to-β Conformational Transition of Alzheimer’s Aβ-(1-42) Peptide in Aqueous Media is Reversible: A Step by Step Conformational Analysis Suggests the Location of β Conformation Seeding. ChemBioChem 2006, 7, 257−267. 20. Crescenzi, O.; Tomaselli, S.; Guerrini, R.; Salvadori, S.; D’Ursi, A. M.; Temussi, P. A.; Picone, D. Solution Structure of the Alzheimer Amyloid β-peptide (1-42) in An Apolar Microenvironment. Eur. J. Biochem. 2002, 269, 5642−5648. 21. Talafous, J.; Marcinowski, K. J.; Klopman, G.; Zagorski, M. G. Solution Structure of Residues 1-28 of the Amyloid β-Peptide. Biochemistry 1994, 33, 7788−7796. 22. Janssen J. C.; Beck J. A.; Campbell T. A.; Dickinson A.; Fox N. C.; Harvey R. J.; Houlden, H.; Rossor, M. N. J. C. Early Onset Familial Alzheimer's Disease: Mutation Frequency in 31 Families. Neurology 2003, 60, 235–239. 23. Tomidokoro, Y.; Rostagno, A.; Neubert, T. A.; Lu, Y.; Rebeck, G. W.; Frangione, B.; Greenberg, S. M.; Ghiso, J. Iowa Variant of Familial Alzheimer’s Disease: Accumulation of Posttranslationally Modified AbetaD23N in Parenchymal and Cerebrovascular Amyloid Deposits. Am. J. Pathol. 2010, 176, 1841−1854. 24. Maat-Schieman, M.; Roos, R.; van Duinen, S. Hereditary cerebral haemorrhage with amyloidosis-Dutch type. Neuropathology 2005, 25, 288−297. 25. Bugiani, O.; Giaccone, G.; Rossi, G.; Mangieri, M.; Capobianco, R.; Morbin, M.; et al. Hereditary cerebral hemorrhage with amyloidosis associated with the E693K mutation of APP. Arch. Neurol. 2010, 67, 987−995. 26. Miravalle, L.; Tokuda, T.; Chiarle, R.; Giaccone, G.; Bugiani, O.; Tagliavini, F.; Frangione, B.; and Ghiso, J. Substitutions at Codon 22 of Alzheimer’s Abeta Peptide Induce Diverse Conformational Changes and Apoptotic Effects in Human Cerebral Endothelial Cells. J. Biol. Chem. 2000, 275, 27110−27116. 27. Kamino, K.; Orr, H. T.; Payami, H.; Wijsman, E. M.; Alonso, M. E.; Pulst, S. M.; et al. Linkage and Mutational Analysis of Familial Alzheimer Disease Kind Reds for the APP Gene Region. Am. J. Hum. Genet. 1992, 51, 998−1014. 28. Murakami, K.; Irie, K.; Morimoto, A.; Ohigashi, H.; Shindo, M.; Nagao, M.; Shimizu, T.; Shirasawa, T. Neurotoxicity and Physicochemical Properties of Abeta Mutant



Peptides from Cerebral Amyloid Angiopathy: Implication for the Pathogenesis of Cerebral Amyloid Angiopathy and Alzheimer's Disease. J Biol Chem. 2003 278, 46179-46187. 29. Nilsberth, C.; Westlind-Danielsson, A.; Eckman, C. B.; Condron, M. M.; Axelman, K.; Forsell, C.; et al. The ’Arctic’ APP Mutation (E693G) Causes Alzheimer’s Disease by Enhanced Abeta Protofibril Formation. Nat. Neurosci. 2001, 4, 887−893. 30. Eisenhauer, P. B.; Johnson, R. J.; Wells, J. M.; Davies, T. A.; and Fine, R. E. Toxicity of Various Amyloid Beta Peptide Species in Cultured Human Blood-Brain Barrier Endothelial Cells: Increased Toxicity of Dutch-type Mutant. J. Neurosci. Res. 2000, 60, 804−810. 31. Tycko, R.; Sciarretta, K. L.; Orgel, J. P. R. O.; Meredith, S. C. Evidence for Novel βSheet Structures in Iowa Mutant βAmyloid Fibrils. Biochemistry 2009, 48, 6072−6084. 32. Hendriks, L.; van Duijn, C. M.; Cras, P.; Cruts, M.; Van Hul, W.; van Harskamp, F.; Warren, A.; McInnis, M. G.; Antonarakis, S. E.; Martin, J. J.; et al. Presenile Dementia and Cerebral Haemorrhage Linked to a Mutation at Codon 692 of the βAmyloid Precursor Protein Gene. Nat. Genet. 1992, 1, 218−221. 33. Pritam Kumar, P.; Abhaysinha, S. P.; Priyam, P.; Hetalkumar, P. Mutation-based Structural Modification and Dynamics Study of Amyloid Beta Peptide (1–42): An inSilico-Based Analysis to Cognize the Mechanism of Aggregation. Genom Data. 2016, 7, 189–194. 34. Yu-Shan, Lin.; Vijay, S. P. Effects of Familial Mutations on the Monomer Structure of Aβ42. Biophys J. 2012, 103, L47–L49. 35. Krone, M. G.; Baumketner, A.; Bernstein, S. L.; Wyttenbach, T.; Lazo, N. D.; Teplow, D. B.; Bowers, M. T.; Shea, J. E. Effects of Familial Alzheimer's Disease Mutations on the Folding Nucleation of the Amyloid Beta-Protein. J Mol Biol. 2008, 381, 221-228. 36. Baumketner, A.; Krone, M. G.; Shea, J. E. Role of the Familial Dutch Mutation E22Q in the Folding and Aggregation of the 15-28 Fragment of the Alzheimer AmyloidBeta Protein. Proc Natl Acad Sci U S A. 2008, 105, 6027-6032. 37. Phan Minh, T.; Man Hoang, V.; Phuong, H. N.; Chin-Kun, Hu.; Mai Suan, Li. Effect of Taiwan Mutation (D7H) on Structures of Amyloid-β Peptides: Replica Exchange Molecular Dynamics Study. J. Phys. Chem. B 2014, 118, 8972–8981. 38. Luhrs, T.; Ritter, C.; Adrian, M.; Riek-Loher, D.; Bohrmann, B.; Dobeli, H.; Schubert, D.; Riek, R. Proc.Natl.Acad.Sci. Usa 2005, 102, 17342-17347. 39. Nose, S.; Klein, M. L. Constant Pressure Molecular Dynamics for Molecular Systems. Mol. Phys. 1983, 50, 1055−1076. 40. Parrinello, M.; Rahman, A. Polymorphic Transitions in Single Crystals: A New Molecular Dynamics Method. J. Appl. Phys. 1981, 52, 7182−7190. 41. Bussi, G.; Donadio, D.; Parrinello, M. Canonical Sampling through Velocity Rescaling. J. Chem. Phys. 2007, 126, 14101−14107. 42. Berendsen, H. J. C.; van der Spoel, D.; van Drunen, R. GROMACS: A MessagePassing Parallel Molecular Dynamics Implementation. Comput. Phys. Commun. 1995, 91, 43−56.


Page 18 of 27

Page 19 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


43. Lindahl, E.; Hess, B.; van der Spoel, D. GROMACS 3.0: A Package for Molecular Simulation and Trajectory Analysis. J. Mol. Model. 2001, 7, 306−317. 44. Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable Molecular Simulation. J. Chem. Theory Comput. 2008, 4, 435−447. 45. Reman, K. S.; Neharika, G. C.; Suman, C.; Arnab, M. Mechanism of Unfolding of Human Prion Protein. J. Phys. Chem. B 2017, 121, 550−564. 46. Bonomi, M.; Branduardi, D.; Bussi, G.; Camilloni, C.; Provasi, D.; Raiteri, P.; Donadio, D.; Marinelli, F.; Pietrucci, F.; Broglia, R. A.; Parrinello, M. Plumed: A Portable Plugin for Free-Energy Calculations with Molecular Dynamics. Comput. Phys. Commun. 2009, 180, 1961−1972. 47. Kumar, S.; Rosenberg, J. M.; Bouzida, D.; Swendsen, R. H.; Kollman, P. A. The Weighted Histogram Analysis Method for Free-Energy Calculations on Biomolecules. I. The Method. J. Comput. Chem., 1992, 13, 1011-1021. 48. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A. V.; Bloino, J.; Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J. V.; Izmaylov, A. F.; Sonnenberg, J. L.; WilliamsYoung, D.; Ding, F.; Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.; Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.; Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J. J.; Brothers, E. N.; Kudin, K. N.; Staroverov, V. N.; Keith, T. A.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J. Gaussian 16, Revision B.01, Gaussian, Inc., Wallingford CT, 2016. 49. Aleksandr, V. M.; Christopher, J. C.; Truhlar, D. G. Universal Solvation Model Based on Solute Electron Density and on a Continuum Model of the Solvent Defined by the Bulk Dielectric Constant and Atomic Surface Tensions. J. Phys. Chem. B 2009, 113, 6378–6396.



Wild type A21G D23N E22G E22K E22Q

5.0 4.5 4.0

Free Energy(kcal/mol)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 27

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -200

-150

-100

-50

0

50

100

150

200

Average Psi()

Figure 1. Free energy surface for the conversion of α-helix to β-sheet conformation in all the cases.


Page 21 of 27


240 220 200

Relative Energy(kcal/mol)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


180 160 140 120 100 80 60 40 20 0 -20 -100

-50

0

50

100

150

200

Average Psi()

Figure 2. The calculated relative energies for conformations of different psi angles at B3LYPD/6-31G* level of theory for all the cases.



Helix

Intermediate (coil)

Wild Type

A21G


Page 22 of 27

Beta Sheet

Page 23 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


D23N

E22G



E22K

E22Q

Figure 3. The structure of three minima on the energy surface of amyloid beta dimers in the case of wild type, A21G, D23N, E22G, E22K and E22Q.


Page 24 of 27

Page 25 of 27


16 14 12

Hydrogen Bonds

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


10 8 6 4 2 0 -40

-20

0

20

40

60

80

100

120

140

160

180

200

Average Psi()

Figure 4. The average number of hydrogen bonds in amyloid dimers of wild type, A21G, D23N, E22G, E22K and E22Q.




10

Intermolecular Hydrogen Bonds

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 27

8

6

4

2

0 -40

-20

0

20

40

60

80

100

120

140

160

180

Average Psi()

Figure 5. The average number of intermolecular hydrogen bonds in amyloid dimers of wild type, A21G, D23N, E22G, E22K and E22Q.


Page 27 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 6. The calculated relative energies for conformations of different psi angles at PM7 level of theory for all the cases.


Effect of Familial Mutations on the Interconversion of Alpha Helix to

Recommend Documents