Membrane Interactions of hIAPP Monomer and Oligomer with Lipid

Jun 6, 2017 - ... where r(0) is the reference position of lipid phosphorus atoms, r(t) is the position of lipid phosphorus atoms in the determined tim...
1 downloads 11 Views 5MB Size
Subscriber access provided by Binghamton University | Libraries

Article

Membrane Interactions of hIAPP Monomer and Oligomer with Lipid Membranes by Molecular Dynamics Simulations Mingzhen Zhang, Baiping Ren, Yonglan Liu, Guizhao Liang, Yan Sun, Lijian Xu, and Jie Zheng ACS Chem. Neurosci., Just Accepted Manuscript • Publication Date (Web): 06 Jun 2017 Downloaded from http://pubs.acs.org on June 8, 2017

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 free 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 accessible to all readers and 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.

ACS Chemical Neuroscience 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

Membrane Interactions of hIAPP Monomer and Oligomer with Lipid Membranes by Molecular Dynamics Simulations Mingzhen Zhang1,3ζ, Baiping Ren3ζ, Yonglan Liu3, Guizhao Liang3, Yan Sun2, Lijian Xu1,3*, and Jie Zheng3* 1

College of Life Sciences and Chemistry Hunan University of Technology, Zhuzhou 412007, China 2

Department of Biochemical Engineering and Key Laboratory of Systems Bioengineering of the Ministry of Education School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China 3

ζ

Department of Chemical and Biomolecular Engineering The University of Akron, Akron, Ohio 44325

The authors contribute equally to this work

1 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

Abstract Interaction of human islet amyloid polypeptide (hIAPP) peptides with cell membrane is crucial for the understanding of amyloid toxicity associated with Type II diabetes (T2D). While it is known that the hIAPP-membrane interactions are considered to promote hIAPP aggregation into fibrils and induce membrane disruption, the membrane-induced conformation, orientation, aggregation, and adsorption behaviors of hIAPP peptides have not been well understood at atomic level. Herein, we perform all-atom explicit-water molecular dynamics (MD) simulations to study the adsorption, orientation, and surface interaction of hIAPP aggregates with different sizes (monomer to tetramer) and conformations (monomer with α-helix and tetramer with βsheet-rich U-turn) upon adsorption on the lipid bilayers composed of both pure zwitterionic POPC (1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine) and mixed anionic POPC/POPE (1palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine) (3:1) lipids. MD simulation results show that hIAPP monomer with α-helical conformation and hIAPP pentamer with β-sheet conformation can adsorb on both POPC and POPC/POPE bilayers via a preferential orientation of N-terminal residues facing toward the bilayer surface. The hIAPP aggregates show stronger interactions with mixed POPC/POPE lipids than pure POPC lipids, consistent with experimental observation that hIAPP adsorption and fibrililation are enhanced on mixed lipid bilayers. While electrostatic interactions are main attractive forces to drive the hIAPP aggregates to adsorb on both bilayers, the introduction of the more hydrophilic head groups of POPE lipids further promote the formation of the interfacial hydrogen bonds. Complement to our previous studies of hIAPP aggregates in bulk solution, this computational work increases our knowledge about the mechanism of amyloid peptide-membrane interactions that is central to the understanding the progression of all amyloid diseases. Keywords: hIAPP, lipid bilayer, amyloid aggregation, cytotoxicity, membrane absorption

2 ACS Paragon Plus Environment

Page 2 of 27

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

ACS Chemical Neuroscience

Introduction Human islet amyloid polypeptide (hIAPP, also known as amylin), a 37-residue peptide hormone, is synthesized and co-secreted with insulin from the pancreatic β-cells 1-5. hIAPP plays many biological functions in islet functions, including carbohydrate metabolism, insulin secretion, bone resorption, gastric clearance, and blood glucose regulation 6-10. However, under some unknown physiological conditions, hIAPP peptides can misfold and self-assemble into aggregates of different structures and sizes. These aggregates, particularly small oligomers, are believed to contribute to the loss of pancreatic β-cells responsible for the pathology of type II diabetes (T2D). In a broader viewpoint, other early amyloid oligomers formed by Aβ 11 and αsynuclein 12 have also been considered as the primary toxic species to induce neurodegeneration in different tissues associated with Alzheimer and Parkinson diseases13-18. The hIAPP-mediated toxicity mechanisms are very complex and involve different pathways to induce β-cell death, including the formation of excessive reactive oxygen species, the increase of end aoplasmic reticulum stress, the initiation of inflammatory response, and membrane disruption. Among them, a number of studies have shown that membrane disruption induced by hIAPP plays an important role in cell toxicity 5, 19-23. hIAPP oligomers can directly interact with cell membranes, disrupting the integrity, permeability, and functions of cell membranes, leading to ionic homeostasis, changes of signaling pathways, oxidative injury, and cell death ultimately 5, 24-29. It has been also reported that introduction of other small molecules including organic molecules, polymers, and metal ions into natural and artificial membranes may alter the functions and properties of the membranes 30-33. However, it remains unclear how hIAPP oligomers interact with cell membranes. While hIAPP monomers are proved as inert species due to the poor correlation with islet cell viability, the interaction of hIAPP monomers with cell membranes may facilitate not only the aggregation of hIAPP monomers and the formation of toxic hIAPP oligomers, but also the structural transition of the adsorbed species towards the β-sheet structures 34, 35. As a result, membrane-bound hIAPP oligomers will accelerate amyloid formation and induce membrane leakage 36. Thus, characterizing the membrane structures and interactions of early hIAPP monomers and oligomers is critical for a better understanding of amyloid aggregation and toxicity mechanisms. Islet β-cell has very complex membrane components containing many different lipids of LPC (lysophosphatidylcholine), SPH1 (sphingomyelin1), SPH2 (sphingomyelin2), PC (phosphatidylcholine), LPE (lysophosphatidylethanolamine), PE (phosphatidylethanolamine), PI (phosphatidylinositol), PS (phosphatidylserne), PG (phosphatidylglycerol), PA (phosphatidic acid), and CL (cardiolipin) 37. Among them, PC and PE are the two major lipids in the islet cell membranes, occupying >60% of total lipids 37. Membrane interactions of hIAPP peptides depend on both lipid components/structures/dynamics and hIAPP structures/sizes 38, 39. At the pathological environment, hIAPP carries four positive charges due to positively-charged residues (K1, R11 and H18) and N-terminus. A number of experimental studies have shown that anionic lipid bilayers containing PS or PG lipids promote hIAPP adsorption and accelerate hIAPP fibrililation 16, 25, 40. Molecular dynamics simulations have also showed that anionic POPG lipid bilayers favor strong hIAPP-lipid interactions to promote hIAPP adsorption 41, 42. It is not surprising that the attractive electrostatic interactions between anionic lipids and cationic hIAPP promote hIAPP absorption and aggregation on the anionic membrane. However, it should be noted that the anionic lipids are usually concentrated in the cytosolic side of the plasma 3 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 4 of 27

membranes, not in the extracellular side that mainly contains neutral lipids 20. This suggests that the interactions of hIAPP with those neutral lipids in the extracellular side are more biologically relevant to understand the mechanisms of hIAPP aggregation and toxicity. Atomistic simulation of a heterogeneous mixture model of islet β-cell membrane incorporating all possible lipids is almost infeasible for current computational capabilities. In this work, to model a realistic islet β-cell membrane, we have chosen the most abundant PC and PE lipids to construct two lipid bilayers, i.e. pure POPC bilayer and mixed POPC/POPE (3:1) bilayer. Two distinct conformations of full-length hIAPP, i.e., α-helix and β-sheet-rich U-turn, are selected to present the initial structures of hIAPP monomers and oligomers. Experimental and simulation studies of hIAPP have revealed that upon binding to cell membranes, hIAPP monomers adopt two kinked α-helix motifs at residue 7-17 and 21-28 and one 310 α-helix at residue 33-3522, and then gradually aggregate into β-sheet-rich oligomers, protofibrils, and mature fibrils 43. Herein, we employ all-atom explicit-solvent molecular dynamics (MD) simulations to study the peptide adsorption, conformational dynamics, and surface interaction of full-length hIAPP monomer with α-helix conformation and small hIAPP pentamer with β-sheet conformation on two model POPC and POPC/POPE (3:1) bilayers. The presence of different lipid components and hIAPP aggregates allows to investigate the combinatorial effects of lipid compositions/structures/dynamics and hIAPP sizes/conformations on the adsorption, orientation, and interactions of hIAPP on lipid bilayers. Simulation results showed that hIAPP monomer and pentamer could stably associate with both POPC and POPC/POPE bilayers, but via specific Nterminal residues to interact with the lipids. Moreover, hIAPP aggregates interacted more strongly with and located more closely to mixed POPC/POPG lipids than pure POPC lipids, resulting from cooperative hIAPP-lipid interactions from hydrogen bonds and salt bridges. Combined with our previous structural study of hIAPP aggregation in solution, this study provides parallel insights into the interactions of hIAPP with lipid bilayers, which is useful for future development of therapeutic agents and approaches to effectively disrupt hIAPP adsorption, aggregation, and surface interaction on the membrane. Results and Discussion Surface Characterization of hIAPP Monomers and Oligomers Since the external surface of hIAPP aggregates is major interacting site for surface adsorption of hIAPP aggregates on a lipid bilayer, we first characterized their surface properties in terms of hydrophilic, hydrophobic, and charged characteristics. Numerous studies have shown that hIAPP monomer mainly adopts random coil in solution, but upon binding to cell membrane, it undergoes structural transition to α-helix structure. As shown in Figure 1, two typical αhelices of hIAPP monomer formed by residues C7–V17 and N21–S28 are connected by a small turn of H18–S20, resulting in a two kinked helix motifs 22. hIAPP monomer has two opposite external surfaces, defined as S1 and S2, both of which possess high probability to interact with lipid bilayers. Due to the different side chain orientations, S1 surface consists of 42.6% hydrophobic, 48.2% hydrophilic, and 9.2% positively-charged surface areas, while S2 surface consists of 20.1% hydrophobic, 66.7% hydrophilic, and 13.2% positively-charged surface areas. Thus, S1 surface is more hydrophobic, and S2 surface is more hydrophilic. Meanwhile, a hIAPP pentamer was selected as a representative hIAPP oligomer for studying its membrane interactions, since our previous works have demonstrated that hIAPP pentamer presents high structural stability with a well preserved U-turn β-sheet-rich conformation 44. hIAPP pentamer was constructed by 4 ACS Paragon Plus Environment

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

ACS Chemical Neuroscience

stacking five U-turn hIAPP monomers on the top of each other in an in-register manner with the inter-peptide distance of 4.7 Ǻ. As shown in Figure 1, hIAPP pentamer has three external surfaces, i.e., (i) a N-terminal surface consists of the N-terminal β-sheet strands (residue 1-19) and contains 16.4% hydrophobic, 55.2% hydrophilic, and 28.4% positively-charged surface areas; (ii) a cross-section U-shaped surface consists of both N-terminal and C-terminal residues (residue 1-37) and contains 30.6% hydrophobic, 56.9% hydrophilic, and 12.5% positivelycharged surface areas; and (iii) a C-terminal surfaces consists of the C-terminal β-sheet strands (residue 26-37) and contains 37.2% hydrophobic and 62.8% hydrophilic surface areas without the charged surface area. Characterization of POPC and POPC/POPE Membranes The islet β cell membranes mainly contain PC and PE lipids, occupying >60% of a total of membrane components37. A number of studied have shown that hIAPP can aggregate and adsorb on both model POPC and POPC/POPE bilayers45, 46. We firstly constructed pure POPC and mixed POPC/POPE (3:1) bilayers and performed 100-ns MD simulations for both bilayers in explicit water solution at 0.35 M ionic strength (0.015 M NaCl and 0.015 M CaCl2) (Figure 1). Visual inspection of MD trajectories showed that both pure POPC and POPC/POPE bilayers experienced structural relaxations and adjustments in the initial 20 ns and then achieved equilibrium state in the next 80-ns. Figure 2a-b shows the position probability distributions for different characteristic groups of PO4, CH3, N(CH3)3 for POPC lipid and PO4, CH3, NH3 for POPE lipid. In both lipid bilayers, all groups exhibited symmetrical distributions relative to the center of the bilayer (z=0) along the bilayer normal (z-axis), indicating that the lipids in the upper and lower leaflets remain the almost same arraignments and no obvious structural disturbances were observed. The peak locations of PO4 and N groups in POPC/POPE bilayer were slightly narrower than those in POPC bilayer, leading to the smaller area per lipid in POPC/POPE bilayer (62.8±0.1 Ǻ2) than POPC bilayer (65.2±0.1 Ǻ2) consistent with the experimental values of ~65 Ǻ2 for POPC 47. So, the presence of small headgroup of NH3 in POPE lipids favors more compact lateral packing. The separation distance between PO4 groups in the upper and lower leaflets can be used to estimate the thickness of lipid bilayers. As shown in Figure 2c, both pure POPC and POPC/POPE bilayers presented stable membrane thickness with very minor fluctuations throughout the entire simulations, where POPC/POPE bilayer (38.1±0.2 Ǻ) had a slightly smaller thickness than POPC bilayer (39.5±0.2 Ǻ). Since the structure and dynamics of lipid bilayers are critical for regulating the function and activity of cell membranes, we introduced the two key parameters of lipid diffusion coefficient and membrane order parameter to characterize the dynamics and structures of lipids, respectively. Lipid diffusivity coefficient is calculated from the analysis of mean-square displacement (MSD) in the Einstein relation in two dimensions, , where r(0) is the reference position of lipid phosphorus atoms, r(t) is the position of lipid phosphorus atoms in the determined time t, and the bracket < > denote a time and particle number average. Large lipid diffusion coefficients indicate the high fluidity of lipid bilayer, and vice versa. The calculated diffusion coefficients were ~5.40*10-7 cm2 s–1 for POPC and ~6.09*10-7 cm2 s–1 for POPC/POPE, respectively, indicating that pure POPC has the slightly lower fluidity than mixed POPC/POPE bilayers. Membrane order parameter was calculated by Px=0.5*(3*cos2(τ)-1) 48, where the τ represents the angle between the positional vector connecting lipid carbon chain (between C2 and the C29-C39 center in POPC lipid) and the bilayer normal. Px reflects the lipid alignment 5 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

relative to the membrane normal, i.e. Px=1 corresponds to perfectly ordered lipids, while the Px=0 corresponds to randomly mixed lipids. Figure 2d presents the average membrane order parameters as a function of simulation time. Pure POPC and mixed POPC/POPE lipid bilayers possessed the steady membrane order parameters along the simulations, with the averaged order parameter of 0.74±0.03 for POPC bilayer and 0.72±0.04 for POPC/POPE bilayers, suggesting that (i) both lipid bilayers contain the ordered lipids and (ii) mixing POPC and POPE lipids does not introduce obvious mismatch to lipid packing and structure, consistent with our previous work 48 . Both equilibrated bilayers are used for the subsequent MD simulations of membrane interactions with hIAPP. Meanwhile, we also calculated the probability distribution profiles of Ca2+, Na+, and Cl- along the bilayer normal. As shown in Figure 2e-f, Ca2+ displayed a strong binding tendency to PO43− headgroups over other Na+ and Cl- ions. The highly populated Ca2+ binding sites were indicated by two sharp broad peaks at z of ~±18 Ǻ near the top and bottom leaflets. A small amount of Na+ ions were also observed to be weakly adsorbed on the bilayer surface. In contrast, Cl- ions were almost distributed in bulk solution and did not favor membrane adsorption. Another possibility is that strong interactions of Ca2+ with lipid can block the binding sites, further preventing other cations from getting into the bilayer surface. This observation also implies that Ca2+ may play an important role in modulating the membrane binding of hIAPP peptides 49. Adsorption of hIAPP Monomers on Lipid Bilayers Membrane interactions play a central role in the aggregation and toxicity of hIAPP peptides . It is generally believed that membrane binding of hIAPP monomers is an initial step for facilitating amyloid formation on cell membranes 50. While previous works have shown that hIAPP monomers mainly adopt α-helix conformation on negatively-charged lipids 51, it is still unclear whether different lipid compositions would have different influences on the binding conformation and affinity of hIAPP monomers. Considering that hIAPP monomer has two distinct external surfaces (S1 and S2) to interact with each lipid bilayer, a total of four simulation systems were constructed to study the adsorption of hIAPP monomers on both bilayers, as summarized in Table 1. MD trajectories showed that when the S1 surface was employed as an initial interacting face to directly contact with membranes, hIAPP monomer failed to establish steady and strong interactions with both lipid bilayers. Consequently, hIAPP monomer flew away from POPC bilayer at 106 ns and POPC/POPE bilayer at 82 ns, respectively. Further analysis of membrane interactions of hIAPP-S1 becomes unnecessary and thus will not be discussed here. In the case of the S2 surface facing towards the lipid surface, hIAPP monomer consistently absorbed on both lipid bilayers throughout 200 ns simulations (Figure 3a-b). The time-dependent residue distance profiles for residue K1, Q10, S20, T30, and Y37 relative to the bilayer surfaces (PO4 group at Z = ~ 20 Ǻ) were monitored and recorded in Figure 3c-d. The five residues represent different locations of distinct structural motifs of hIAPP monomer, i.e. Ntermini (K1), N-terminal α-helix (Q10), C-terminal α-helix (S20), 310 helix structure (T30), and Ctermini (Y37). The distances of these five residues relative to bilayer surface were monitored to reflect the interaction extents among different structural motifs. For MonoS2POPC systems, K1 and Q10 could maintain their initial contacts with POPC bilayer with very small distance fluctuations and some N-terminal residues tended to insert into POPC bilayer, indicating that Nterminal residues of hIAPP monomer from the S2 surface enable to maintain steady interactions with POPC lipids. Three S20, T30 and Y37 residues experienced large distance fluctuations, particularly C-terminal residues drifted away from the POPC bilayer with a large separation 20

6 ACS Paragon Plus Environment

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

ACS Chemical Neuroscience

distance of ~15 Ǻ. Simulation observation is consistent with previous experiments that hIAPP has week interactions with pure POPC membranes 50. Different from weak binding of hIAPP monomer on POPC bilayer, all five residues experienced small distance fluctuations and maintained their distances within