European Journal of Medicinal Chemistry 92 (2015) 876e881
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Resveratrol interferes with the aggregation of membrane-bound human-IAPP: A molecular dynamics study Fabio Lolicato a, b, Antonio Raudino a, Danilo Milardi c, *, Carmelo La Rosa a, * a
Department of Chemical Sciences, University of Catania, Viale A. Doria 6, Catania I-95125, Italy Department of Physics, Tampere University of Technology, P.O. Box 692, FI-33101 Tampere, Finland c Organizzativa e di Supporto di Catania, Viale A. Doria 6, Catania I-95125, Italy Istituto di Biostrutture e Bioimmagini, CNR, Unita b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 June 2014 Received in revised form 22 January 2015 Accepted 23 January 2015 Available online 24 January 2015
Amyloid aggregation of islet amyloid polypeptide (IAPP) in pancreatic tissues is a typical feature of type 2 diabetes mellitus. Resveratrol, a natural product extensively studied for its wide range of biological effects, has been shown to inhibit IAPP aggregation. However, the mechanism by which resveratrol inhibits IAPP aggregation is still far from complete elucidation. Now, an increasing knowledge of the mechanism of amyloid toxicity shifts the target of research towards the development of compounds which can prevent amyloid-mediated membrane damage rather than merely inhibit fiber formation. In this study we used all atom molecular dynamics to investigate the interaction of resveratrol with full-length human IAPP in a negatively charged membrane environment. Our results show that the presence of resveratrol induces the formation of secondary structures (sheets and helices) by inserting in a hydrophobic pocket between the interaction surface of two IAPP molecules in aqueous solution. On the other hand, resveratrol significantly perturbs the interaction of IAPP with negatively charged membranes by anchoring specific hydrophobic regions (23FGA25 and 32VGS34) of the peptide and forming a stable 1:2 IAPP:resveratrol complex at the water/membrane interphase. © 2015 Elsevier Masson SAS. All rights reserved.
Keywords: Molecular dynamics Atomistic simulation Type II diabetes Membrane Resveratrol
1. Introduction An increasingly large number of proteins are known to lack a defined structure under physiological conditions [1]. Due to their delicate biological tasks, these intrinsically disordered proteins (IDPs) are involved in a large number of human diseases [2] and represent one of the most attractive (and challenging) drug target of the last decade. Important examples of IDPs which form pathogenic amyloid aggregates in vivo include: the Ab peptide of Alzheimer's disease; a-synuclein, responsible for Lewy body formation in Parkinson's disease and IAPP, which is the protein component of type II diabetes-associated islet amyloid [3] Upon interacting with other cytosolic partners such as proteins or membranes, IDPs can partly (mis)fold into (dys) functional/toxic conformations [4e6]: therefore, amyloid formation in vivo may follow quite different routes from those observed in dilute aqueous solutions and studies of IDPs assembly in heterogeneous water/membrane environment are expected to provide more significant advances in our
* Corresponding authors. E-mail addresses: [email protected] (D. Milardi), [email protected] (C. La Rosa). http://dx.doi.org/10.1016/j.ejmech.2015.01.047 0223-5234/© 2015 Elsevier Masson SAS. All rights reserved.
understanding of these pathogenic mechanisms [7,8]. One of the most extensively characterized amyloid/membrane paradigm is IAPP, whose interactions with lipid vesicles have been studied by different groups [9e17]. According to these reports IAPP, as many other IDPs, may induce membrane damage, which is believed to be a central event in type II diabetes pathogenesis [10]. As a consequence, inhibiting the growth of IAPP fibrils and their potential to destabilize membranes is considered as a promising therapeutic strategy toward type II diabetes [18]. Polyphenols are a family of small molecules which have an inhibitory effect on amyloid formation [19]. In particular resveratrol (trans-3,5,40 -trihydroxystilben), has evidenced significant beneficial effects in patients with T2DM [20]. Moreover, several studies have demonstrated that resveratrol may inhibit IAPP amyloid aggregation in the presence of negatively charged membranes [16,21]. Molecular Dynamics simulations have also demonstrated that resveratrol inhibits the aggregation of IAPP segment 22e27 by interfering with the intersheet side-chain stacking driven by aromatic rings [22] Nevertheless, there is still need of a more detailed structural characterization of the dynamics of membrane/peptide assemblies to fully understand the mechanism by which
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
resveratrol inhibits IAPP fibrillogenesis and toxicity. Despite the wealth of information on IAPP-induced changes in membrane properties and on the structural rearrangements of membranebound IAPP, a thorough description of the conformational dynamics of the IAPP/membrane systems remains unapproachable by conventional experimental techniques. In this light, in silico methods could play an irreplaceable role in a thorough description of the atomistic implications of experimental evidences related to IAPP/membrane/inhibitor assemblies, as well as in generating de novo predictions useful in drug design. To address these issues, here we employ fully atomistic molecular dynamics simulations to study the aggregation of unstructured full-length IAPP in water/negatively charged membrane in the presence of resveratrol. Our results evidence that resveratrol inhibits conformational transitions of membrane-bound IAPP by anchoring specific hydrophobic regions (23FGA25 and 32VGS34) of the peptide. Moreover, the IAPP/resveratrol assembly has a limited diffusion dynamics onto the membrane surface which may damp the proteineprotein association process. 2. Results 2.1. Molecular dynamics of hIAPP in aqueous solution: the effect of resveratrol A wealth of experimental data has evidenced that hIAPP has a remarkable tendency to adopt a b-sheet rich secondary structure in electrolyte solution. Notably, previous NMR and CD experiments have evidenced that the N-terminal domain of hIAPP exhibits a conformational propensity to adopt a-helical structures leaving the other residues unstructured [23e25]. However, it must be noted that experimental techniques may only provide an average description of peptide conformational preferences and cannot distinguish oligomers from monomers. By contrast, replica exchange MD simulations have identified two coexixsting families of monomeric hIAPP conformers in solution: an helix-coil structure and a b-hairpin conformation [26]. Here, in agreement with those previous simulations, 400 ns of atomistic MD showed that hIAPP undergoes a conformational transition from ahelix (starting structure) to a b-sheet arrangement coexisting with a transient helix/random coil structure as shown in Fig. 1. Next, we explored the self-assembling dynamics of two hIAPP molecules during 400 ns of MD simulation. It was observed the
877
formation of a linear aggregate containing one molecule almost completely unstructured and the second one with large a-helix content (Fig. 2, upper panel). It can be noted, that an U-shaped unstructured molecule is in contact with a second one by its C and N termini. Resveratrol significantly influenced the dynamics of hIAPP conformational preferences and aggregation. The presence of resveratrol at 1:1 polyphenol/peptide molar ratio decreases the number of inter-peptide contacts in hIAPP dimeric assembly (see fig. S1). In Fig. 2 lower panel, it is also evidenced how resveratrol significantly modifies the structure and the average contact maps of dimeric hIAPP. Contact maps show as resveratrol favor both intra and inter side chain contacts of two IAPP molecules. In order to better understand the effect of resveratrol on the hIAPP selfassembling, we have investigated the interactions of resveratrol with monomeric hIAPP. Fig 3 reports the final conformation of hIAPP monomer in the presence of resveratrol (molar ratio 1:1) after 200 ns of MD simulation. hIAPP/resveratrol assembly preferentially adopts a b-sheet rich U-shaped conformation. The helical structure of hIAPP unfolds due to the interaction of resveratrol with the unstructured N- and C-terminus domains of the peptide. Resveratrol weakly interacts with hIAPP: only 10% hIAPP configurations were observed to be in contact with resveratrol during 200 ns of MD simulation. But at higher hIAPP/resveratrol concentration a different behavior was evidenced. Peptides aggregate after the first 14 ns the residues Leu12, Ser19, Phe23, Gly24, Ala25 of one molecule are in contact with residues Ala8 and Thr9 of the second molecule. After 172 ns, these residues form a stable pocket where one molecule of resveratrol may be accommodated (fig. S2). A second resveratrol molecule has weak and non-specific interactions with the dimeric hIAPP assembly. This trend has been observed in repeated simulations starting from different initial configurations randomly chosen. The conformational preferences of the dimeric hIAPP assembly are influenced by resveratrol as well. Fig. 4 reports the secondary structure evolutions vs time of the dimer in the presence and in the absence of resveratrol. Resveratrol induces higher b-sheet content; on the contrary, pure hIAPP exhibits higher random coil and a-helix contents. 2.2. Interaction of hIAPP with negatively charged POPG membranes First we explored the interaction of an isolated hIAPP molecule with palmitoyloleylphosphatidylglycerol (POPG, negatively
Fig. 1. Left panel: secondary structures evolution over 400 ns of atomistic simulation of hIAPP in water solution. Color code: white, random coil; yellow, b-sheet; green, turn; pink, a-helix; red, 3-10 helix. Right panel: cartoon representation. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
878
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
Fig. 2. Cartoon representation of the final structure (left) and of the residueeresidue contact map (right) as averaged matrix over the whole trajectory after 200 ns of atomistic simulation. Upper panel: hIAPP dimer aggregate in water; lower panel: the same dimeric structure in presence of resveratrol.
charged) lipid bilayer starting the simulations (200 ns) with the protein freely floating in the electrolytic aqueous phase. POPG model membranes were selected to mimic hIAPP-membranes interactions because they have saturated hydrocarbon chains: this is an important prerequisite to ensure that the lipid bilayer is in a liquid crystalline state at 300 K. Moreover, the negative charge lipid molecule are supposed to rapidly promote hIAPP/lipid interactions thus decreasing the simulation time. The hIAPP/membrane surface interaction occurred after 31.5 ns, mainly involving the hydrophilic Ser34 side chain. After 58 ns of simulation a hydrophilic domain of hIAPP (Asn 21 and Asn 22) was involved (Fig. 5).
After binding to the lipid bilayer, hIAPP slowly diffuses parallel to the surface with a diffusion coefficient about 23 Å2/ns. The transfer of hIAPP from the water phase to the membrane surface does not imply any appreciable conformational change and, unexpectedly, the membrane-bound peptide undergoes wider conformational fluctuations with respect to those found in the bulk electrolyte solution as evidenced by the RMSD values that increase from 20 to 37 Å. As a further control, we performed additional 50 ns of simulation of resveratrol on the membrane surface. In order to avoid undesired artifacts due to periodic boundary conditions we set the simulations with resveratrol on both sides of the bilayer.
Fig. 3. Interaction of resveratrol-amylin after 200 ns of atomistic simulation.
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
879
Fig. 4. Time line of amylin dimer (left panel) in water and in presence of resveratrol (right panel). Color code: white, random coil; yellow, b-sheet; green, turn; pink, a-helix; red, 310 helix. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Resveratrol irreversibly binds to the membrane surface, showing no tendency to aggregate. Moreover, it preferentially visits the membrane region at the interfacial phosphate polar head forming a hydrogen bond between the phosphate oxygen and a resveratrol hydroxil. 2.3. Interaction of membrane-bound hIAPP with resveratrol Next, we investigated the influence of resveratrol on the dynamics of hIAPP and membrane by 200 ns MD simulation. After 6 ns hIAPP binds to the membrane interacting with resveratrol by stacking contacts involving Phe23. After 15 ns the protein is spread on the membrane surface. At 24 ns of simulation a second resveratrol molecule interacts with the hIAPP C-terminus and with residues Val32, Gly33 and Ser34. Within 200 ns of following simulation, hIAPP remains stable on the membrane surface and the structure of the hIAPP/resveratrol aggregate does not change. It is interesting to note that, while in the absence of resveratrol, the interaction of hIAPP with the membrane surface does not induce any conformational transition, in the presence of resveratrol we observe a decrease in b-sheet structures and structuring of the Nterminus into a-helix (see Fig. 6). 2.4. Cluster analysis To provide a faithful representation of the trajectories and conformational ensembles, the MD simulations were also analyzed by clustering the sampled configurations of the last 20 ns of each trajectory where aggregation processes are going to completion. These data were also compared with the configurations of the last 100 ns. The results show that the number of clusters drastically decreases in the last 20 ns of each simulation. This, probably, means that after the first 100 nanoseconds the structures were still not stabilized. Furthermore, the presence of resveratrol and, mainly, of the membrane causes a smaller flexibility of the protein structure shown by the calculated number of clusters (see Table 1). It is important to note that the number of clusters was significantly dropped in these systems especially if compared with the amylinwater systems. This assumption can also be confirmed by analyzing the Root Mean Square Fluctuations (RMSF) of the three systems in water (see Fig. 7). The presence of resveratrol and, especially, the presence of the membrane stiffens the hIAPP conformation due to the strong interactions with nearby systems (see Fig. 6 for the number of
Protein-membrane contacts). The graph, also, shows a similar trend in the NNFGAIL region (Residues 21e27) for the systems containing resveratrol (red and green lines), indicating that this specific portion of protein plays a crucial role when it interacts with other molecules and then, probably, in the amylineamylin aggregation. The most stable structures obtained from cluster analysis applied to the last 20 ns of the MD trajectories are shown in the Supplementary Information (see Figure S2). 3. Conclusion Isolated hIAPP in 0.1 M electrolyte solution prefers a conformational arrangement showing coexisting b-sheet and a-helix structures. Upon aggregation hIAPP peptides assume a different conformational arrangement that depends on the aggregation number and solvent. Focusing on the dimeric structure alone, we observed that dimers form aggregates with one structured molecule (sheet and helix) and one unstructured. The presence of resveratrol promotes the formation of structured secondary structures (sheet and helix) and induces the formation of a hydrophobic pocket where a molecule of resveratrol is hosted. Next, hIAPP binds to negatively charged membranes, in agreement with experimental measurements [27], preferentially
Fig. 5. Snapshot of hIAPP conformation at POPG surface after 200 ns of simulation.
880
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
Fig. 6. Left panel: Snapshot taken at different time (a, 6 ns; b, 15 ns; c, 24 ns; d, 200 ns) of hIAPP interacting with POPG membrane containing resveratrol (amylin:resveratrol 1:5). For the sake of clarity only the interacting resveratrol with the membrane was shown. Right panel: Number of resveratrol-membrane (blue line), protein-membrane (black line) and protein-resveratrol (red line) contacts. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
through the hydrophilic side chains 21 and 22, while resveratrol in contact with negatively charged membranes irreversibly forms hydrogen bond with the oxygen of the phosphates. A strikingly behavior is found when hIAPP contemporarily interacts with negatively charged lipid membranes and resveratrol. hIAPP in the presence of resveratrol does not interact with the polar heads of phospholipid, rather it forms a stable complex with resveratrol at the water/membrane interphase. Moreover resveratrol, induces hIAPP to assume secondary structures very similar to those the amylin-resveratrol complex forms in electrolyte solution. 4. Simulation setup An extensive set of MD simulations at 300 K were performed using the GROMACS [28] simulation package. The initial protein structure used was the NMR structure of hIAPP bound to sodium dodecylsufate (SDS) micelles [29] (pdb ID: 2KB8). It was first energy-minimized using the method of steepest descents. Preequilibration was performed in the NPT ensemble for 20 ns. This step was followed by a 400 ns simulation to stabilize the structure in solution. The monomeric structure thus obtained was used as the starting configuration of all production MD simulations. The Geometry optimization and the partial charges calculation of resveratrol molecule were carried out using Automated Topology Builder (ATB) 2.0 web server [30] by choosing HF/STO-3G as basis set [31]. Pre-quilibration was performed in the NPT ensemble for 20 ns and after it was followed by a 50 ns simulation to stabilize the structure in solution. As far as the lipid bilayer is concerned, a preequilibrated 40 ns containing 108 lipid of POPG membrane [32,33] hydrated by water molecules was used to simulate a negatively charged model membrane. Since POPG show chiral
center, according literature data [34], a racemic mixture was used in order to have a lipid bilayer characterized by neutral chirality. This assembly was the starting point to build the POPG/Resveratrol (5:1) bilayer in a box of 108 64 64 Å. Pre-equilibration was performed in NPT ensamble for 50 ns in order to stabilize the position of resveratrol into the POPG lipid bilayer. The atomistic simulations were carried out using Gromacs 4.6.5 [32]. The GROMOS 53A6 force field [35,36] was used. This forcefield has been shown to perform well in simulating proteineprotein and membrane-protein systems [37,16]. The SPC (simple point charge) water model [38] was used. The time step was set to 2 fs and the temperature was kept constant using the V-rescale algorithm [39], with a time constant of 0.1 ps. Periodic boundary conditions were applied and the Parrinello-Rahman algorithm [40] was applied for isotropic and semi-isotropic pressure coupling (1 bar). For the electrostatic interactions the particle mesh Ewald method [41] was employed with a real space cut-off of 1.2 nm, cubic interpolation (4th order), and a direct sum tolerance of 10-5. All systems were hydrated in the presence of 0.1 M NaCl. Overall
Table 1 Number of clusters of the sampled configurations (analyzed system numbered in the first column) of the last 100 and 20 ns of each trajectory. Structures/ number of molecules
Amylin Resveratrol POPG N clusters N clusters Average (last 100 ns) (last 20 ns) RMSD
1 2 3 4 5
1 2 1 2 1
0 0 1 2 20
0 0 0 0 108
71 394 82 171 7
19 70 7 10 2
0.42 2.49 0.30 1.16 0.22
Fig. 7. RMSF analysis of monomeric systems in pure water (black line), in the presence of one molecule of resveratrol (green line) and in contact with a negatively charged membrane (red line). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
charge neutrality was preserved by adding Cl counterions. For all system a pre-equilibration was performed in NPT ensemble for 20 ns followed by a molecular dynamics simulation at 300 K for 200 ns. To simulate the hIAPP assembly into dimer, the final structure obtained after a 200 ns MD simulation of the monomer was replicated and then placed into the solvent box whereas in membrane-protein systems the monomer, obtained in the previous way, was initially positioned above the membrane in the water phase. For the analysis and visualization we used the VMD [42] (visual Molecular Dynamics) package, the DSSP [43] (Definition of Secondary Structure of Protein) program and the GROMACS [23] analysis tools.
[18]
[19]
[20]
[21]
Acknowledgments
[22]
The authors are grateful to Martina Pannuzzo and Robert Lagana for fruitful discussions and support. We acknowledge University of Catania and PRIN 2010/2011, Cod. prog.: 2010L96H3K for financial support.
[23]
Appendix A. Supplementary data
[24]
[25]
Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ejmech.2015.01.047.
[26]
References
[27]
[1] P.E. Wright, H.J. Dyson, Intrinsically unstructured proteins: reassessing the protein structure-function paradigm, J. Mol. Biol. 293 (1999) 321e331. [2] V.N. Uversky, C.J. Oldfield, A.K. Dunker, Intrinsically disordered proteins in human diseases: introducing the D2 concept, Annu. Rev. Biophys. 37 (2008), 15e246. [3] A. Abedini, D.P. Raleigh, A critical assessment of the role of helical intermediates in amyloid formation by natively unfolded proteins and polypeptides, Prot. Eng. Des. Selec 22 (2009) 453e459. [4] A.D. Snow, T.N. Wight, Proteoglycans in the pathogenesis of Alzheimer's disease and other amyloidoses, Neurobiol. Aging 10 (1989) 481e497. [5] J.D. Sipe, Amyloidosis, Crit. Rev. Clin. Lab. Sci. 31 (1994) 325e354. [6] C.M. Dobson, Protein misfolding, evolution and disease, Trends Biochem. Soc. 24 (1999) 329e332. [7] C. La Rosa, D. Milardi, D.M. Grasso, M.P. Verbeet, G.W. Canters, L. Sportelli, R. Guzzi, A model for the thermal unfolding of amicyanin, E. Biophys. J. 30 (2002) 559e570. [8] C. La Rosa, D. Milardi, S. Fasone, D. Grasso, The effects of scan rate and protein concentration on DSC thermograms of bovine superoxide dismutase, Thermochim. Acta 265 (1995) 163e175. [9] J.D. Knight, A.D. Miranker, Phospholipid catalysis of diabetic amyloid assembly, J. Mol. Biol. 341 (2004) 1175e1187. [10] S.A. Jayasinghe, R. Langen, Tracking fiber formation in human islet amyloid polypeptide with automated 2D-IR spectroscopy, Biochemistry 44 (2005) 12113e12119. [11] S.A. Jayasinghe, R. Langen, Membrane interaction of islet amyloid polypeptide, Biochim. Biophys. Acta Biomembr. 1768 (2007) 2002e2009. [12] J.R. Brender, S. Salamekh, A. Ramamoorthy, Membrane disruption and early events in the aggregation of the diabetes related peptide IAPP from a molecular perspective, Acc. Chem. Res. 45 (2012) 454e462. [13] M.F.M. Sciacca, M. Pappalardo, D. Milardi, D.M. Grasso, C. La Rosa, Calciumactivated membrane interaction of the islet amyloid polypeptide: implications in the pathogenesis of type II diabetes mellitus, Arc. Biochem. Biophys. 477 (2008) 291e298. [14] M.F.M. Sciacca, D. Milardi, G.M.L. Messina, G. Marletta, J.R. Brender, A. Ramamoorthy, C. La Rosa, Cations as switches of amyloid-mediated membrane disruption mechanisms: calcium and IAPP, Biophys. J. 104 (2013) 173e184. [15] D. Milardi, M.F.M. Sciacca, M. Pappalardo, D.M. Grasso, C. La Rosa, The role of aromatic side-chains in amyloid growth and membrane interaction of the islet amyloid polypeptide fragment LANFLVH, Eur. Biophys. J. 40 (2011) 1e12. [16] M. Pannuzzo, A. Raudino, D. Milardi, C. La Rosa, M. Karttunen, a-Helical structures drive early stages of self-assembly of amyloidogenic amyloid polypeptide aggregate formation in membranes, Sci. Rep. 3 (2013) 2781. [17] M.F.M. Sciacca, M. Pappalardo, F. Attanasio, D. Milardi, C. La Rosa, D.M. Grasso,
[28] [29] [30]
[31]
[32]
[33]
[34]
[35] [36]
[37]
[38]
[39] [40] [41] [42] [43]
881
Are fibril growth and membrane damage linked processes? an experimental and computational study of IAPP12e18 and IAPP21e27 peptides, New. J. Chem. 34 (2009) 200e207. F. Evers, C. Jeworrek, S. Tiemeyer, K. Weise, D. Sellin, M. Paulus, B. Struth, M. Tolan, R. Winter, Elucidating the mechanism of lipid membrane-induced IAPP fibrillogenesis and its inhibition by the red wine compound resveratrol: a synchrotron X-ray reflectivity study, J. Am. Chem. Soc. 131 (2009) 9516e9521. Y. Porat, A. Abramowitz, E. Gazit, Inhibition of amyloid fibril formation by polyphenols: structural similarity and aromatic interactions as a common inhibition mechanism, Chem. Biol. Drug Des. 67 (2006) 27e37. K.P. Goh, H.Y. Lee, D.P. Lau, W. Supaat, Y.H. Chan, A.F.Y. Koh, Effects of resveratrol in patients with type 2 diabetes mellitus on skeletal muscle SIRT1 expression and energy expenditure, Int. J. Sport Nutr. Exerc. Metab. 24 (2014) 2e13. M.F.M. Sciaccaa, R. Chillemi, S. Sciuto, M. Pappalardo, C. La Rosa, D. Grasso, D. Milardi, Interactions of two O-phosphorylresveratrol derivatives with model membranes, Arc. Biochem. Biophys. 521 (2012) 111e116. P. Jiang, W. Li, J.-E. Shea, Y. Mu, Resveratrol inhibits the formation of multiplelayered b-sheet oligomers of the human islet amyloid polypeptide segment, Biophys. J. 22 (2011) 1550e1558. I.T. Yonemoto, G.J.A. Kroon, H.J. Dyson, W.E. Balch, J.W. Kelly, Amylin proprotein processing generates progressively more amyloidogenic peptides that initially sample the helical state, Biochemistry 47 (2008) 9900e9910. S.M. Vaiana, R.B. Best, W.M. Yau, W.A. Eaton, J. Hofrichter, Evidence for a partially structured state of the amylin monomer, Biophys. J. 97 (2009) 2948e2957. R. Soong, J.R. Brender, P.M. MacDonald, A. Ramamoorthy, Association of highly compact type II diabetes related islet amyloid polypeptide intermediate species at physiological temperature revealed by diffusion NMR spectroscopy, J. Am. Chem. Soc. 131 (2009) 7079e7085. N.F. Dupuis, C. Wu, J.E. Shea, M.T. Bowers, The amyloid formation mechanism in human IAPP: dimers have b-strand monomermonomer interfaces, J. Am. Chem. Soc. 113 (2011) 7240e7243. S. Scalisi, M.F.M. Sciacca, G. Zhavnerko, D.M. Grasso, G. Marletta, C. La Rosa, Self-Assembling pathway of HiApp fibrils within lipid bilayers, ChemBioChem 11 (2010) 1856e1859. E. Lindahl, B. Hess, D. van der Spoel, GROMACS 3.0: a package for molecular simulation and trajectory analysis, J. Mol. Model 7 (2001) 306e317. S.M. Patil, S. Xu, S.R. Sheftic, A.T. Alexandrescu, Dynamic alpha-helix structure of micelle-bound human amylin, J. Biol. Chem. 284 (2009) 11982e11991. A.K. Malde, L. Zuo, M. Breeze, M. Stroet, D. Poger, P.C. Nair, C. Oostenbrink, A.E. Mark, An automated force field topology builder (ATB) and repository: version 1.0, J. Chem. Theory Comput. 12 (2011) 4026e4037. M.J.S. Dewar, E.G. Zoebisch, E.F. Healy, J.J.P. Stewart, The development and use of quantum-mechanical molecular-models 0.76. AM1-A new general-purpose quantum-mechanical molecular model, J. Am. Chem. Soc. 107 (1985) 3902e3909. B. Hess, C. Kutzner, D. Van der Spoel, E. Lindahl, GROMACS 4: algorithms for highly efficient, load-balanced, and scalable molecular simulation, J. Chem. Theory Comput. 4 (2008) 435e447. I. Pascher, S. Sundell, K. Harlos, H. Eibl, Conformation and packing properties of membrane lipids: the crystal structure of sodium dimyristoylphosphatidylglycerol, Biochem. Biophys. Acta 896 (1987) 77e88. g, A.A. Gurtovenko, I. Vattulainen, M. Karttunen, Atomic-scale W. Zhao, T. Ro structure and electrostatics of anionic aalmitoyloleoylphosphatidylglycerol lipid bilayers with Naþ counterions, Biophys. J. 92 (2007) 1114e1124. E.A. Cino, W.-Y. Choy, M. Karttunen, Conformational biases of linear motifs, J. Phys. Chem. B 117 (2013) 15943e15957. E.A. Cino, W.-Y. Choy, M. Karttunen, Comparison of secondary structure formation using 10 different force fields in microsecond molecular dynamics simulations, J. Chem. Theory Comput. 8 (2012) 2725e2740. S.R. Gerben, J.A. Lemkul, A.M. Brown, D.R. Bevan, Comparing atomistic molecular mechanics force fields for a difficult target: a case study on the Alzheimer's amyloid b-peptide, J. Biomol. Struct. Dyn. 31 (2013) 1e16. H.J.C. Berendsen, J.P.M. Postma, van W.F. Gunsteren, J. Hermans, Interaction models for water in relation to protein hydration, in: B. Pullman (Ed.), Intermolecular Forces, D. Reidel, Dordrecht, Holland, 1981, pp. 331e342. G. Bussi, D. Donadio, M. Parrinello, Canonical sampling through velocity rescaling, J. Chem. Phys. 126 (2007) 014101. M. Parrinello, A. Rahman, Polymorphic transitions in single crystals: a new molecular dynamics method, J. Appl. Phys. 52 (1981) 7182e7190. T. Darden, D. York, L. Pedersen, Particle mesh Ewald: an Nlog(N) method for Ewald sums in large systems, J. Chem. Phys. 98 (1993) 10089e10092. W. Humphrey, A. Dalke, K. Schulten, VMD: visual molecular dynamics, J. Mol. Graph. 14 (1996) 33e38. W. Kabsch, C. Sander, Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features, Biopolymers 22 (1983) 2577e2637.
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Resveratrol interferes with the aggregation of membrane-bound human-IAPP: A molecular dynamics study Fabio Lolicato a, b, Antonio Raudino a, Danilo Milardi c, *, Carmelo La Rosa a, * a
Department of Chemical Sciences, University of Catania, Viale A. Doria 6, Catania I-95125, Italy Department of Physics, Tampere University of Technology, P.O. Box 692, FI-33101 Tampere, Finland c Organizzativa e di Supporto di Catania, Viale A. Doria 6, Catania I-95125, Italy Istituto di Biostrutture e Bioimmagini, CNR, Unita b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 June 2014 Received in revised form 22 January 2015 Accepted 23 January 2015 Available online 24 January 2015
Amyloid aggregation of islet amyloid polypeptide (IAPP) in pancreatic tissues is a typical feature of type 2 diabetes mellitus. Resveratrol, a natural product extensively studied for its wide range of biological effects, has been shown to inhibit IAPP aggregation. However, the mechanism by which resveratrol inhibits IAPP aggregation is still far from complete elucidation. Now, an increasing knowledge of the mechanism of amyloid toxicity shifts the target of research towards the development of compounds which can prevent amyloid-mediated membrane damage rather than merely inhibit fiber formation. In this study we used all atom molecular dynamics to investigate the interaction of resveratrol with full-length human IAPP in a negatively charged membrane environment. Our results show that the presence of resveratrol induces the formation of secondary structures (sheets and helices) by inserting in a hydrophobic pocket between the interaction surface of two IAPP molecules in aqueous solution. On the other hand, resveratrol significantly perturbs the interaction of IAPP with negatively charged membranes by anchoring specific hydrophobic regions (23FGA25 and 32VGS34) of the peptide and forming a stable 1:2 IAPP:resveratrol complex at the water/membrane interphase. © 2015 Elsevier Masson SAS. All rights reserved.
Keywords: Molecular dynamics Atomistic simulation Type II diabetes Membrane Resveratrol
1. Introduction An increasingly large number of proteins are known to lack a defined structure under physiological conditions [1]. Due to their delicate biological tasks, these intrinsically disordered proteins (IDPs) are involved in a large number of human diseases [2] and represent one of the most attractive (and challenging) drug target of the last decade. Important examples of IDPs which form pathogenic amyloid aggregates in vivo include: the Ab peptide of Alzheimer's disease; a-synuclein, responsible for Lewy body formation in Parkinson's disease and IAPP, which is the protein component of type II diabetes-associated islet amyloid [3] Upon interacting with other cytosolic partners such as proteins or membranes, IDPs can partly (mis)fold into (dys) functional/toxic conformations [4e6]: therefore, amyloid formation in vivo may follow quite different routes from those observed in dilute aqueous solutions and studies of IDPs assembly in heterogeneous water/membrane environment are expected to provide more significant advances in our
* Corresponding authors. E-mail addresses: [email protected] (D. Milardi), [email protected] (C. La Rosa). http://dx.doi.org/10.1016/j.ejmech.2015.01.047 0223-5234/© 2015 Elsevier Masson SAS. All rights reserved.
understanding of these pathogenic mechanisms [7,8]. One of the most extensively characterized amyloid/membrane paradigm is IAPP, whose interactions with lipid vesicles have been studied by different groups [9e17]. According to these reports IAPP, as many other IDPs, may induce membrane damage, which is believed to be a central event in type II diabetes pathogenesis [10]. As a consequence, inhibiting the growth of IAPP fibrils and their potential to destabilize membranes is considered as a promising therapeutic strategy toward type II diabetes [18]. Polyphenols are a family of small molecules which have an inhibitory effect on amyloid formation [19]. In particular resveratrol (trans-3,5,40 -trihydroxystilben), has evidenced significant beneficial effects in patients with T2DM [20]. Moreover, several studies have demonstrated that resveratrol may inhibit IAPP amyloid aggregation in the presence of negatively charged membranes [16,21]. Molecular Dynamics simulations have also demonstrated that resveratrol inhibits the aggregation of IAPP segment 22e27 by interfering with the intersheet side-chain stacking driven by aromatic rings [22] Nevertheless, there is still need of a more detailed structural characterization of the dynamics of membrane/peptide assemblies to fully understand the mechanism by which
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
resveratrol inhibits IAPP fibrillogenesis and toxicity. Despite the wealth of information on IAPP-induced changes in membrane properties and on the structural rearrangements of membranebound IAPP, a thorough description of the conformational dynamics of the IAPP/membrane systems remains unapproachable by conventional experimental techniques. In this light, in silico methods could play an irreplaceable role in a thorough description of the atomistic implications of experimental evidences related to IAPP/membrane/inhibitor assemblies, as well as in generating de novo predictions useful in drug design. To address these issues, here we employ fully atomistic molecular dynamics simulations to study the aggregation of unstructured full-length IAPP in water/negatively charged membrane in the presence of resveratrol. Our results evidence that resveratrol inhibits conformational transitions of membrane-bound IAPP by anchoring specific hydrophobic regions (23FGA25 and 32VGS34) of the peptide. Moreover, the IAPP/resveratrol assembly has a limited diffusion dynamics onto the membrane surface which may damp the proteineprotein association process. 2. Results 2.1. Molecular dynamics of hIAPP in aqueous solution: the effect of resveratrol A wealth of experimental data has evidenced that hIAPP has a remarkable tendency to adopt a b-sheet rich secondary structure in electrolyte solution. Notably, previous NMR and CD experiments have evidenced that the N-terminal domain of hIAPP exhibits a conformational propensity to adopt a-helical structures leaving the other residues unstructured [23e25]. However, it must be noted that experimental techniques may only provide an average description of peptide conformational preferences and cannot distinguish oligomers from monomers. By contrast, replica exchange MD simulations have identified two coexixsting families of monomeric hIAPP conformers in solution: an helix-coil structure and a b-hairpin conformation [26]. Here, in agreement with those previous simulations, 400 ns of atomistic MD showed that hIAPP undergoes a conformational transition from ahelix (starting structure) to a b-sheet arrangement coexisting with a transient helix/random coil structure as shown in Fig. 1. Next, we explored the self-assembling dynamics of two hIAPP molecules during 400 ns of MD simulation. It was observed the
877
formation of a linear aggregate containing one molecule almost completely unstructured and the second one with large a-helix content (Fig. 2, upper panel). It can be noted, that an U-shaped unstructured molecule is in contact with a second one by its C and N termini. Resveratrol significantly influenced the dynamics of hIAPP conformational preferences and aggregation. The presence of resveratrol at 1:1 polyphenol/peptide molar ratio decreases the number of inter-peptide contacts in hIAPP dimeric assembly (see fig. S1). In Fig. 2 lower panel, it is also evidenced how resveratrol significantly modifies the structure and the average contact maps of dimeric hIAPP. Contact maps show as resveratrol favor both intra and inter side chain contacts of two IAPP molecules. In order to better understand the effect of resveratrol on the hIAPP selfassembling, we have investigated the interactions of resveratrol with monomeric hIAPP. Fig 3 reports the final conformation of hIAPP monomer in the presence of resveratrol (molar ratio 1:1) after 200 ns of MD simulation. hIAPP/resveratrol assembly preferentially adopts a b-sheet rich U-shaped conformation. The helical structure of hIAPP unfolds due to the interaction of resveratrol with the unstructured N- and C-terminus domains of the peptide. Resveratrol weakly interacts with hIAPP: only 10% hIAPP configurations were observed to be in contact with resveratrol during 200 ns of MD simulation. But at higher hIAPP/resveratrol concentration a different behavior was evidenced. Peptides aggregate after the first 14 ns the residues Leu12, Ser19, Phe23, Gly24, Ala25 of one molecule are in contact with residues Ala8 and Thr9 of the second molecule. After 172 ns, these residues form a stable pocket where one molecule of resveratrol may be accommodated (fig. S2). A second resveratrol molecule has weak and non-specific interactions with the dimeric hIAPP assembly. This trend has been observed in repeated simulations starting from different initial configurations randomly chosen. The conformational preferences of the dimeric hIAPP assembly are influenced by resveratrol as well. Fig. 4 reports the secondary structure evolutions vs time of the dimer in the presence and in the absence of resveratrol. Resveratrol induces higher b-sheet content; on the contrary, pure hIAPP exhibits higher random coil and a-helix contents. 2.2. Interaction of hIAPP with negatively charged POPG membranes First we explored the interaction of an isolated hIAPP molecule with palmitoyloleylphosphatidylglycerol (POPG, negatively
Fig. 1. Left panel: secondary structures evolution over 400 ns of atomistic simulation of hIAPP in water solution. Color code: white, random coil; yellow, b-sheet; green, turn; pink, a-helix; red, 3-10 helix. Right panel: cartoon representation. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
878
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
Fig. 2. Cartoon representation of the final structure (left) and of the residueeresidue contact map (right) as averaged matrix over the whole trajectory after 200 ns of atomistic simulation. Upper panel: hIAPP dimer aggregate in water; lower panel: the same dimeric structure in presence of resveratrol.
charged) lipid bilayer starting the simulations (200 ns) with the protein freely floating in the electrolytic aqueous phase. POPG model membranes were selected to mimic hIAPP-membranes interactions because they have saturated hydrocarbon chains: this is an important prerequisite to ensure that the lipid bilayer is in a liquid crystalline state at 300 K. Moreover, the negative charge lipid molecule are supposed to rapidly promote hIAPP/lipid interactions thus decreasing the simulation time. The hIAPP/membrane surface interaction occurred after 31.5 ns, mainly involving the hydrophilic Ser34 side chain. After 58 ns of simulation a hydrophilic domain of hIAPP (Asn 21 and Asn 22) was involved (Fig. 5).
After binding to the lipid bilayer, hIAPP slowly diffuses parallel to the surface with a diffusion coefficient about 23 Å2/ns. The transfer of hIAPP from the water phase to the membrane surface does not imply any appreciable conformational change and, unexpectedly, the membrane-bound peptide undergoes wider conformational fluctuations with respect to those found in the bulk electrolyte solution as evidenced by the RMSD values that increase from 20 to 37 Å. As a further control, we performed additional 50 ns of simulation of resveratrol on the membrane surface. In order to avoid undesired artifacts due to periodic boundary conditions we set the simulations with resveratrol on both sides of the bilayer.
Fig. 3. Interaction of resveratrol-amylin after 200 ns of atomistic simulation.
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
879
Fig. 4. Time line of amylin dimer (left panel) in water and in presence of resveratrol (right panel). Color code: white, random coil; yellow, b-sheet; green, turn; pink, a-helix; red, 310 helix. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Resveratrol irreversibly binds to the membrane surface, showing no tendency to aggregate. Moreover, it preferentially visits the membrane region at the interfacial phosphate polar head forming a hydrogen bond between the phosphate oxygen and a resveratrol hydroxil. 2.3. Interaction of membrane-bound hIAPP with resveratrol Next, we investigated the influence of resveratrol on the dynamics of hIAPP and membrane by 200 ns MD simulation. After 6 ns hIAPP binds to the membrane interacting with resveratrol by stacking contacts involving Phe23. After 15 ns the protein is spread on the membrane surface. At 24 ns of simulation a second resveratrol molecule interacts with the hIAPP C-terminus and with residues Val32, Gly33 and Ser34. Within 200 ns of following simulation, hIAPP remains stable on the membrane surface and the structure of the hIAPP/resveratrol aggregate does not change. It is interesting to note that, while in the absence of resveratrol, the interaction of hIAPP with the membrane surface does not induce any conformational transition, in the presence of resveratrol we observe a decrease in b-sheet structures and structuring of the Nterminus into a-helix (see Fig. 6). 2.4. Cluster analysis To provide a faithful representation of the trajectories and conformational ensembles, the MD simulations were also analyzed by clustering the sampled configurations of the last 20 ns of each trajectory where aggregation processes are going to completion. These data were also compared with the configurations of the last 100 ns. The results show that the number of clusters drastically decreases in the last 20 ns of each simulation. This, probably, means that after the first 100 nanoseconds the structures were still not stabilized. Furthermore, the presence of resveratrol and, mainly, of the membrane causes a smaller flexibility of the protein structure shown by the calculated number of clusters (see Table 1). It is important to note that the number of clusters was significantly dropped in these systems especially if compared with the amylinwater systems. This assumption can also be confirmed by analyzing the Root Mean Square Fluctuations (RMSF) of the three systems in water (see Fig. 7). The presence of resveratrol and, especially, the presence of the membrane stiffens the hIAPP conformation due to the strong interactions with nearby systems (see Fig. 6 for the number of
Protein-membrane contacts). The graph, also, shows a similar trend in the NNFGAIL region (Residues 21e27) for the systems containing resveratrol (red and green lines), indicating that this specific portion of protein plays a crucial role when it interacts with other molecules and then, probably, in the amylineamylin aggregation. The most stable structures obtained from cluster analysis applied to the last 20 ns of the MD trajectories are shown in the Supplementary Information (see Figure S2). 3. Conclusion Isolated hIAPP in 0.1 M electrolyte solution prefers a conformational arrangement showing coexisting b-sheet and a-helix structures. Upon aggregation hIAPP peptides assume a different conformational arrangement that depends on the aggregation number and solvent. Focusing on the dimeric structure alone, we observed that dimers form aggregates with one structured molecule (sheet and helix) and one unstructured. The presence of resveratrol promotes the formation of structured secondary structures (sheet and helix) and induces the formation of a hydrophobic pocket where a molecule of resveratrol is hosted. Next, hIAPP binds to negatively charged membranes, in agreement with experimental measurements [27], preferentially
Fig. 5. Snapshot of hIAPP conformation at POPG surface after 200 ns of simulation.
880
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
Fig. 6. Left panel: Snapshot taken at different time (a, 6 ns; b, 15 ns; c, 24 ns; d, 200 ns) of hIAPP interacting with POPG membrane containing resveratrol (amylin:resveratrol 1:5). For the sake of clarity only the interacting resveratrol with the membrane was shown. Right panel: Number of resveratrol-membrane (blue line), protein-membrane (black line) and protein-resveratrol (red line) contacts. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
through the hydrophilic side chains 21 and 22, while resveratrol in contact with negatively charged membranes irreversibly forms hydrogen bond with the oxygen of the phosphates. A strikingly behavior is found when hIAPP contemporarily interacts with negatively charged lipid membranes and resveratrol. hIAPP in the presence of resveratrol does not interact with the polar heads of phospholipid, rather it forms a stable complex with resveratrol at the water/membrane interphase. Moreover resveratrol, induces hIAPP to assume secondary structures very similar to those the amylin-resveratrol complex forms in electrolyte solution. 4. Simulation setup An extensive set of MD simulations at 300 K were performed using the GROMACS [28] simulation package. The initial protein structure used was the NMR structure of hIAPP bound to sodium dodecylsufate (SDS) micelles [29] (pdb ID: 2KB8). It was first energy-minimized using the method of steepest descents. Preequilibration was performed in the NPT ensemble for 20 ns. This step was followed by a 400 ns simulation to stabilize the structure in solution. The monomeric structure thus obtained was used as the starting configuration of all production MD simulations. The Geometry optimization and the partial charges calculation of resveratrol molecule were carried out using Automated Topology Builder (ATB) 2.0 web server [30] by choosing HF/STO-3G as basis set [31]. Pre-quilibration was performed in the NPT ensemble for 20 ns and after it was followed by a 50 ns simulation to stabilize the structure in solution. As far as the lipid bilayer is concerned, a preequilibrated 40 ns containing 108 lipid of POPG membrane [32,33] hydrated by water molecules was used to simulate a negatively charged model membrane. Since POPG show chiral
center, according literature data [34], a racemic mixture was used in order to have a lipid bilayer characterized by neutral chirality. This assembly was the starting point to build the POPG/Resveratrol (5:1) bilayer in a box of 108 64 64 Å. Pre-equilibration was performed in NPT ensamble for 50 ns in order to stabilize the position of resveratrol into the POPG lipid bilayer. The atomistic simulations were carried out using Gromacs 4.6.5 [32]. The GROMOS 53A6 force field [35,36] was used. This forcefield has been shown to perform well in simulating proteineprotein and membrane-protein systems [37,16]. The SPC (simple point charge) water model [38] was used. The time step was set to 2 fs and the temperature was kept constant using the V-rescale algorithm [39], with a time constant of 0.1 ps. Periodic boundary conditions were applied and the Parrinello-Rahman algorithm [40] was applied for isotropic and semi-isotropic pressure coupling (1 bar). For the electrostatic interactions the particle mesh Ewald method [41] was employed with a real space cut-off of 1.2 nm, cubic interpolation (4th order), and a direct sum tolerance of 10-5. All systems were hydrated in the presence of 0.1 M NaCl. Overall
Table 1 Number of clusters of the sampled configurations (analyzed system numbered in the first column) of the last 100 and 20 ns of each trajectory. Structures/ number of molecules
Amylin Resveratrol POPG N clusters N clusters Average (last 100 ns) (last 20 ns) RMSD
1 2 3 4 5
1 2 1 2 1
0 0 1 2 20
0 0 0 0 108
71 394 82 171 7
19 70 7 10 2
0.42 2.49 0.30 1.16 0.22
Fig. 7. RMSF analysis of monomeric systems in pure water (black line), in the presence of one molecule of resveratrol (green line) and in contact with a negatively charged membrane (red line). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
F. Lolicato et al. / European Journal of Medicinal Chemistry 92 (2015) 876e881
charge neutrality was preserved by adding Cl counterions. For all system a pre-equilibration was performed in NPT ensemble for 20 ns followed by a molecular dynamics simulation at 300 K for 200 ns. To simulate the hIAPP assembly into dimer, the final structure obtained after a 200 ns MD simulation of the monomer was replicated and then placed into the solvent box whereas in membrane-protein systems the monomer, obtained in the previous way, was initially positioned above the membrane in the water phase. For the analysis and visualization we used the VMD [42] (visual Molecular Dynamics) package, the DSSP [43] (Definition of Secondary Structure of Protein) program and the GROMACS [23] analysis tools.
[18]
[19]
[20]
[21]
Acknowledgments
[22]
The authors are grateful to Martina Pannuzzo and Robert Lagana for fruitful discussions and support. We acknowledge University of Catania and PRIN 2010/2011, Cod. prog.: 2010L96H3K for financial support.
[23]
Appendix A. Supplementary data
[24]
[25]
Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ejmech.2015.01.047.
[26]
References
[27]
[1] P.E. Wright, H.J. Dyson, Intrinsically unstructured proteins: reassessing the protein structure-function paradigm, J. Mol. Biol. 293 (1999) 321e331. [2] V.N. Uversky, C.J. Oldfield, A.K. Dunker, Intrinsically disordered proteins in human diseases: introducing the D2 concept, Annu. Rev. Biophys. 37 (2008), 15e246. [3] A. Abedini, D.P. Raleigh, A critical assessment of the role of helical intermediates in amyloid formation by natively unfolded proteins and polypeptides, Prot. Eng. Des. Selec 22 (2009) 453e459. [4] A.D. Snow, T.N. Wight, Proteoglycans in the pathogenesis of Alzheimer's disease and other amyloidoses, Neurobiol. Aging 10 (1989) 481e497. [5] J.D. Sipe, Amyloidosis, Crit. Rev. Clin. Lab. Sci. 31 (1994) 325e354. [6] C.M. Dobson, Protein misfolding, evolution and disease, Trends Biochem. Soc. 24 (1999) 329e332. [7] C. La Rosa, D. Milardi, D.M. Grasso, M.P. Verbeet, G.W. Canters, L. Sportelli, R. Guzzi, A model for the thermal unfolding of amicyanin, E. Biophys. J. 30 (2002) 559e570. [8] C. La Rosa, D. Milardi, S. Fasone, D. Grasso, The effects of scan rate and protein concentration on DSC thermograms of bovine superoxide dismutase, Thermochim. Acta 265 (1995) 163e175. [9] J.D. Knight, A.D. Miranker, Phospholipid catalysis of diabetic amyloid assembly, J. Mol. Biol. 341 (2004) 1175e1187. [10] S.A. Jayasinghe, R. Langen, Tracking fiber formation in human islet amyloid polypeptide with automated 2D-IR spectroscopy, Biochemistry 44 (2005) 12113e12119. [11] S.A. Jayasinghe, R. Langen, Membrane interaction of islet amyloid polypeptide, Biochim. Biophys. Acta Biomembr. 1768 (2007) 2002e2009. [12] J.R. Brender, S. Salamekh, A. Ramamoorthy, Membrane disruption and early events in the aggregation of the diabetes related peptide IAPP from a molecular perspective, Acc. Chem. Res. 45 (2012) 454e462. [13] M.F.M. Sciacca, M. Pappalardo, D. Milardi, D.M. Grasso, C. La Rosa, Calciumactivated membrane interaction of the islet amyloid polypeptide: implications in the pathogenesis of type II diabetes mellitus, Arc. Biochem. Biophys. 477 (2008) 291e298. [14] M.F.M. Sciacca, D. Milardi, G.M.L. Messina, G. Marletta, J.R. Brender, A. Ramamoorthy, C. La Rosa, Cations as switches of amyloid-mediated membrane disruption mechanisms: calcium and IAPP, Biophys. J. 104 (2013) 173e184. [15] D. Milardi, M.F.M. Sciacca, M. Pappalardo, D.M. Grasso, C. La Rosa, The role of aromatic side-chains in amyloid growth and membrane interaction of the islet amyloid polypeptide fragment LANFLVH, Eur. Biophys. J. 40 (2011) 1e12. [16] M. Pannuzzo, A. Raudino, D. Milardi, C. La Rosa, M. Karttunen, a-Helical structures drive early stages of self-assembly of amyloidogenic amyloid polypeptide aggregate formation in membranes, Sci. Rep. 3 (2013) 2781. [17] M.F.M. Sciacca, M. Pappalardo, F. Attanasio, D. Milardi, C. La Rosa, D.M. Grasso,
[28] [29] [30]
[31]
[32]
[33]
[34]
[35] [36]
[37]
[38]
[39] [40] [41] [42] [43]
881
Are fibril growth and membrane damage linked processes? an experimental and computational study of IAPP12e18 and IAPP21e27 peptides, New. J. Chem. 34 (2009) 200e207. F. Evers, C. Jeworrek, S. Tiemeyer, K. Weise, D. Sellin, M. Paulus, B. Struth, M. Tolan, R. Winter, Elucidating the mechanism of lipid membrane-induced IAPP fibrillogenesis and its inhibition by the red wine compound resveratrol: a synchrotron X-ray reflectivity study, J. Am. Chem. Soc. 131 (2009) 9516e9521. Y. Porat, A. Abramowitz, E. Gazit, Inhibition of amyloid fibril formation by polyphenols: structural similarity and aromatic interactions as a common inhibition mechanism, Chem. Biol. Drug Des. 67 (2006) 27e37. K.P. Goh, H.Y. Lee, D.P. Lau, W. Supaat, Y.H. Chan, A.F.Y. Koh, Effects of resveratrol in patients with type 2 diabetes mellitus on skeletal muscle SIRT1 expression and energy expenditure, Int. J. Sport Nutr. Exerc. Metab. 24 (2014) 2e13. M.F.M. Sciaccaa, R. Chillemi, S. Sciuto, M. Pappalardo, C. La Rosa, D. Grasso, D. Milardi, Interactions of two O-phosphorylresveratrol derivatives with model membranes, Arc. Biochem. Biophys. 521 (2012) 111e116. P. Jiang, W. Li, J.-E. Shea, Y. Mu, Resveratrol inhibits the formation of multiplelayered b-sheet oligomers of the human islet amyloid polypeptide segment, Biophys. J. 22 (2011) 1550e1558. I.T. Yonemoto, G.J.A. Kroon, H.J. Dyson, W.E. Balch, J.W. Kelly, Amylin proprotein processing generates progressively more amyloidogenic peptides that initially sample the helical state, Biochemistry 47 (2008) 9900e9910. S.M. Vaiana, R.B. Best, W.M. Yau, W.A. Eaton, J. Hofrichter, Evidence for a partially structured state of the amylin monomer, Biophys. J. 97 (2009) 2948e2957. R. Soong, J.R. Brender, P.M. MacDonald, A. Ramamoorthy, Association of highly compact type II diabetes related islet amyloid polypeptide intermediate species at physiological temperature revealed by diffusion NMR spectroscopy, J. Am. Chem. Soc. 131 (2009) 7079e7085. N.F. Dupuis, C. Wu, J.E. Shea, M.T. Bowers, The amyloid formation mechanism in human IAPP: dimers have b-strand monomermonomer interfaces, J. Am. Chem. Soc. 113 (2011) 7240e7243. S. Scalisi, M.F.M. Sciacca, G. Zhavnerko, D.M. Grasso, G. Marletta, C. La Rosa, Self-Assembling pathway of HiApp fibrils within lipid bilayers, ChemBioChem 11 (2010) 1856e1859. E. Lindahl, B. Hess, D. van der Spoel, GROMACS 3.0: a package for molecular simulation and trajectory analysis, J. Mol. Model 7 (2001) 306e317. S.M. Patil, S. Xu, S.R. Sheftic, A.T. Alexandrescu, Dynamic alpha-helix structure of micelle-bound human amylin, J. Biol. Chem. 284 (2009) 11982e11991. A.K. Malde, L. Zuo, M. Breeze, M. Stroet, D. Poger, P.C. Nair, C. Oostenbrink, A.E. Mark, An automated force field topology builder (ATB) and repository: version 1.0, J. Chem. Theory Comput. 12 (2011) 4026e4037. M.J.S. Dewar, E.G. Zoebisch, E.F. Healy, J.J.P. Stewart, The development and use of quantum-mechanical molecular-models 0.76. AM1-A new general-purpose quantum-mechanical molecular model, J. Am. Chem. Soc. 107 (1985) 3902e3909. B. Hess, C. Kutzner, D. Van der Spoel, E. Lindahl, GROMACS 4: algorithms for highly efficient, load-balanced, and scalable molecular simulation, J. Chem. Theory Comput. 4 (2008) 435e447. I. Pascher, S. Sundell, K. Harlos, H. Eibl, Conformation and packing properties of membrane lipids: the crystal structure of sodium dimyristoylphosphatidylglycerol, Biochem. Biophys. Acta 896 (1987) 77e88. g, A.A. Gurtovenko, I. Vattulainen, M. Karttunen, Atomic-scale W. Zhao, T. Ro structure and electrostatics of anionic aalmitoyloleoylphosphatidylglycerol lipid bilayers with Naþ counterions, Biophys. J. 92 (2007) 1114e1124. E.A. Cino, W.-Y. Choy, M. Karttunen, Conformational biases of linear motifs, J. Phys. Chem. B 117 (2013) 15943e15957. E.A. Cino, W.-Y. Choy, M. Karttunen, Comparison of secondary structure formation using 10 different force fields in microsecond molecular dynamics simulations, J. Chem. Theory Comput. 8 (2012) 2725e2740. S.R. Gerben, J.A. Lemkul, A.M. Brown, D.R. Bevan, Comparing atomistic molecular mechanics force fields for a difficult target: a case study on the Alzheimer's amyloid b-peptide, J. Biomol. Struct. Dyn. 31 (2013) 1e16. H.J.C. Berendsen, J.P.M. Postma, van W.F. Gunsteren, J. Hermans, Interaction models for water in relation to protein hydration, in: B. Pullman (Ed.), Intermolecular Forces, D. Reidel, Dordrecht, Holland, 1981, pp. 331e342. G. Bussi, D. Donadio, M. Parrinello, Canonical sampling through velocity rescaling, J. Chem. Phys. 126 (2007) 014101. M. Parrinello, A. Rahman, Polymorphic transitions in single crystals: a new molecular dynamics method, J. Appl. Phys. 52 (1981) 7182e7190. T. Darden, D. York, L. Pedersen, Particle mesh Ewald: an Nlog(N) method for Ewald sums in large systems, J. Chem. Phys. 98 (1993) 10089e10092. W. Humphrey, A. Dalke, K. Schulten, VMD: visual molecular dynamics, J. Mol. Graph. 14 (1996) 33e38. W. Kabsch, C. Sander, Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features, Biopolymers 22 (1983) 2577e2637.
