Molecular Mechanism of Stabilizing the Helical ... - ACS Publications

Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI

Article

Molecular Mechanism on Stabilizing Huntingtin N17 Helical Structure in Micelle Environment Leili Zhang, Hongsuk Kang, Francisco X. Vázquez, Leticia Marisel Toledo-Sherman, Binquan Luan, and Ruhong Zhou J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.7b01476 • Publication Date (Web): 17 Apr 2017 Downloaded from http://pubs.acs.org on April 21, 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.

The Journal of Physical Chemistry B 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

The Journal of Physical Chemistry

Molecular Mechanism on Stabilizing Huntingtin N17 Helical Structure in Micelle Environment Leili Zhang1, Hongsuk Kang1, Francisco X. Vázquez1, Leticia Toledo-Sherman2, Binquan Luan1, and Ruhong Zhou1,* 1. Computational Biology Center, IBM Thomas J. Watson Research Center, Yorktown Heights, New York 10598, USA 2. CHDI Management/CHDI Foundation, Los Angeles, CA 90045, USA

*All correspondence should be addressed to [email protected] (RZ)

1 Environment ACS Paragon Plus


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

Abstract Huntington’s disease is a deadly neurodegenerative disease caused by the fibrilization of huntingtin (HTT) exon-1 protein mutants. Despite extensive efforts over the past decade, much remains unknown about the structures of (mutant) HTT exon-1 and their enigmatic roles in aggregation. Particularly, whether the first 17 residues in the N-terminal (HTT-N17) adopt a helical or a coiled structure remains unclear. Here, with the rigorous study of molecular dynamics simulations,

we

explored

the

most

possible

structures

of

HTT-N17

in

both

dodecylphosphocholine (DPC) micelles and aqueous solution, using three commonly applied force fields (OPLS-AA/L, CHARMM36 and AMBER99sb*-ILDNP) to examine the underlying molecular mechanism and rule out the potential artifacts. We show that local environments are essential for determining the secondary structure of HTT-N17. This is evidenced by the insertion of five hydrophobic residues of HTT-N17 into the DPC micelle which promotes the formation of an amphipathic helix, while such amphipathic helix unfolds quickly in aqueous solution. A relatively low free energy barrier (~3 kcal/mol) for the secondary structure transformation was also observed for all three force fields from their respective folding free energy landscapes, which accounts for possible HTT-N17 conformational changes upon environment shifts such as membrane binding and protein complex aggregation.


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


Introduction Huntington’s disease (HD) is a heredetary neurodegenerative disease caused by a mutational expansion of CAG repeats in the exon-1 region of the huntingtin gene (Htt). The entire Huntingtin exon-1 protein (HTT exon-1) consists of three domains: an amphipathic region made up of the first 17 residues (HTT-N17), a polyglutamine (polyQ) domain, and a polyproline-rich domain composed of 38 residues (HTT-C38). Structure wise, HTT exon-1 and its mutants are intrinsically disordered protein (IDP) that once aggregated can form amyloid-like fibers associated with HD.1 The biological function of the HTT-N17 domain has been explored in previous studies. Along with the polyQ domain in HTT exon-1, HTT-N17 is believed to affect early development of human embryos.2 Nuclear-cytoplasmic distribution of HTT exon-1 is strongly correlated to the toxicity of HTT mutant, where nuclear exclusion is beneficial and nuclear inclusion is toxic.3 HTT-N17 was found to regulate nuclear localization of HTT exon-1 by post-translational phosphorylation of several residues in the domain3. For example, in previous studies, the M8P mutation of HTT-N17 was found to block nuclear entry of the HTT exon-1 protein4. As an amphipathic helix in a hydrophobic environment5, HTT-N17 facilitates the targeting of endoplasmic reticulum (ER)4 and mitochondria membranes6 by the protein. Knockout of HTT-N17 in zebrafish or mice induces HTT aggregation in the nuclei, resulting in a more rapid pathogenic phenotype.7-8 More recently, Crick et. al.9 found that nonfibril aggregates of HTT exon-1 displayed the highest toxicity where HTT-N17 destabilizes the nonfibril intermediates, resulting in more stabilized fibrils. On the other hand, another study shows that HTT-N17 significantly speeds up polyQ aggregation.10 How HTT-N17 exactly regulates the aforementioned nuclear inclusion, membrane targeting and HTT exon-1 aggregation, however,



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

are still not well understood. Therefore, in light of the structural roles in Htt’s biological functions, the structural information of HTT-N17 is critical and invaluable in the field of HD researches. Previous experimental and computational studies indicated that the structure of HTT-N17 in amphipathic environment is mainly helical. For instance, Michalek et. al.

5

used NMR to solve

the structure of HTT-N17 in a DPC micelle (PDB-ID:2LD2) as well as the 2,2,2-trifluoroethanol (TFE) solution (PDB-ID:2LD0), and showed that it adopts a helical structure in hyrdrophobic environments. It was also demonstrated that HTT-N17 is helical when co-crystalized with proteins such as Maltose-Binding Protein (MBP)11-12 and Single Chain Fv Fragment (scFv)13. Also, Côté et. al.14 performed simulations of HTT-N17 on a palmitoyloleoylphosphatidylcholine (POPC) bilayer and achieved 85% helical content using the AMBER99sb*-ILDN force field. However, the standalone HTT-N17 in aqueous solution is speculated to be unstructured or in random coil. For example, solution CD spectra and NMR strongly indicated an unstructured HTT-N17 in water.10 Additionally, several computational research groups have studied the stability of the helical HTT-N17 in aqueous solution. Kelley et. al.15 showed a 36% helical content of HTT-N17 starting from both unstructured state and fully-helical state in water solution using the AMBER03 force field. Rossetti et. al.16 performed temperature replica exchange molecular dynamics (T-REMD) simulations and achieved 29% helical content of HTT-N17 in water using the AMBER99 force field. Binette et. al.17 performed hamiltonian replica exchange molecular dynamics (H-REMD) simulations on HTT-N17 in water and obtained 29.3% helical content with the AMBER99sb*-ILDN force field. Coincidentally, in all of the previous computational work on HTT-N17, some variations of the AMBER force field were selected for study (more below).


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


In this study, we conducted a systematic investigation of the molecular mechansims of HTT-N17 under the micellar and aqueous environments using all-atom molecular dynamics simulations (MD). In order to provide a benchmark and also avoid potential artifacts that arise from one particular force field, we explored three commonly used force fields, namely, OPLS-AA (OPLS-AA/L), AMBER (AMBER99sb*-ILDNP), and CHARMM (CHARMM36) for our simulaitons. AMBER99sb*-ILDNP was chosen over other AMBER variations because it has the least bias towards helix. Also we avoided using CHARMM27 since it was known to cause problems in past studies on IDPs. For example, Côté et. al.18 showed that CHARMM27 also results in significant overestimation of the HTT-N17 helical content in aqueous solution (about 80%). It should be noted that Côté et. al.18 also systematically explored the HTT-N17 peptide insertion mechanism and found that different placements of HTT-N17 can lead to different levels of irreversible destabilization of the helix partly due to the difficulty of helix folding on membranes in the currently achievable time scales. In our current work, we focused on the “stabilization” of the helical content of HTT-N17 by the hydrophobic environment (using DPC micelles) with various widely used force fields. Our results show that in the amphipathic environment, all three force fields behave reasonably well by stabilizing HTT-N17 as an α-helix in a DPC micelle, which is in excellent agreement with the experimental data. As for the case of aqueous environment, these force fields suggest the preference of a disordered coil structure for HTT-N17, albeit the fact that simulations using AMBER exhibit more helical content than the other two force fields. Despite the force field differences, we observe a universal small free energy barrier of secondary structural change of HTT-N17. Our findings clarify the ambiguity of HTT-N17 structures reported in previous works and suggest a possible structural shifting mechanism of HTT-N17 in different environments.



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

Methods Molecular dynamics (MD) simulation We used GROMACS 5.0.5 for all simulations in this study. The OPLS-AA/L (OPLS-AA) force field19 was directly taken from the default GROMACS library. The AMBER99sb*-ILDNP (AMBER) force field20 was obtained from the online library of the GROMACS website (http://www.gromacs.org/). The CHARMM36 (CHARMM) force field21 was obtained from the MacKerell lab’s website (http://mackerell.umaryland.edu/). The united atom lipid force field by Berger et. al.22 was adjusted according to the DPC molecule. We combined Berger’s lipid force field with OPLS-AA and AMBER based on previous studies23-24. DPC micelles (containing 54 DPC molecules) were generated with the CHARMM-GUI website25. We utilized both virtual sites for the protein hydrogens and rigid H-bonds to enable a 4 fs time step for simulations with the OPLS-AA force field. Meanwhile, only the rigid H-bond option was enabled for simulations with the AMBER or CHARMM force fields, to settle for a 2 fs time step. The N-terminus of the protein was capped by –NH3+ and the C-terminus was capped by –COO-. Three initial structures of HTT-N17 (MATLEKLMKAFESLKSF) were prepared with different force field setups: PDB-ID 2LD0, PDB-ID 2LD25 and helix (artificially generated perfect α-helix). Following similar protocols used in our previous studies26-29, all systems were solvated with TIP3P water and neutralized with 100 mM NaCl solution. The initial structures for the micelle simulations are prepared so that HTT-N17 lies horizontally on the DPC micelle with its hydrophobic residues facing towards the micelle and its heavy atoms placed at least a few angstroms away from the DPC micelle. Production runs in isobaric-isothermic ensemble (NPT) were performed following 50000 steps of steepest descent energy minimization and 10 ns of constant volume, constant


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


temperature (NVT) equilibration. Due to different secondary structure decay time (see the time-dependent plots for both helicity (Fig. S1, Fig. S2, and Fig. S3), and radius of gyration (Fig. S4, Fig. S5 and Fig. S6) provided in Supporting Information), data were acquired over 400 ns from each of the micelle simulations and 1.2 µs from each of the solution simulations. Structures were recorded with a 100 ps interval. Secondary structure analyses were performed using the STRIDE algorithm30.

Results HTT-N17 shows largely a helical structure on DPC micelles As mentioned above, the helicity (defined as the percentage of residues being α-helix or 3-10 helix) of HTT-N17 has been disputed ever since the discovery of its importance to the HTT protein.5,

10-13

Previous computational works have addressed this problem from various

aspects14-17, but none actually offers a direct comparison in different environments. Here, we started the simulations from the NMR structures of HTT-N17 (PDB-ID 2LD0, PDB-ID 2LD2) and an artificially generated perfect helix (referred to as 2LD0, 2LD2 and Helix, respectively hereafter) in both aqueous solution and DPC micelles, using three commonly applied force fields, OPLS-AA, AMBER, and CHARMM. The three initial structures used for the simulations are respectively shown in Fig. 1A, Fig. 1B and Fig. 1C. A quick examination of the HTT-N17 sequence shows its amphipathicity from the banded distribution of hydrophobic (bold) and hydrophilic (italic) residues: MATLEKLMKAFESLKSF. How these hydrophobic and hydrophilic residues contribute to its strucural preference, particularly its helical content under different environments, is thus of great importance (and interest) given the complex nature of protein-protein interactions that HTT involves.



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

For the DPC micelles, the HTT-N17 in all simulations converged towards similar helical structures on the micelle surface (Fig. 1G, Fig. 1H and Fig. 1I). To show the orientation and insertion depth of each residue’s side-chain in HTT-N17 on the micelle surface, we calculated the distance between the center of mass (COM) of each residue’s side-chain and the COM of the entire micelle. To account for shape fluctuation of the micelle (e.g. deviation from a spherical one), that distance was further normalized by the distance bewteen the COMs of the HTT-N17 and of the micelle, resulting in a good agreement among all simulations regardless of the starting point and force field used (Fig. 2A). Generally, negative values of normalized distances indicate that corresponding residue side-chains (e.g. hydrophobic ones) were inserted in the micelle, while positive values suggest that corresponding residues (e.g. hydrophilic ones) had their side-chains pointed away from the micelle surface. The illustration of relative positions of each residue with regard to the membrane-water interface is shown in Fig. 2B. The end residues M1, A2 and F17 are found to be more flexible than the rest of the peptide, where secondary structures are not well-defined and RMSFs (root-mean-square fluctuations) are significantly larger due to their end locations (Fig. S7). Excluding these flexible terminal residues, we observed that the L4, L7, A10, F11 and L14 residues inserted into the hydrophobic center of DPC micelles, while the T3, E5, K9, E12 and S16 residues pointed towards the water. The overall orientation thus ensures the maximal insertion of the hydrophobic residues into hydrophobic micelle environment. In all, our simulated HTT-N17 structures agree well with the NMR structures5 and highlight the importance of the amphiphilic nature of the peptide. The detailed interactions between HTT-N17 are further discussed in the section “Insertion patterns of HTT-N17 in DPC micelles”, with additional information on contact number analyses and insertion process analyses.


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


Additioanlly, we calculated the average helicity of HTT-N17 on the micelle surface, as shown in Fig. 1J, Fig. 1K, Fig. 1L and Table 1. If starting from a perfect helix as the reference (Fig. 1C, Fig. S9C), the OPLS-AA force field has a higher chance of forming a helix from residue 3-12. The CHARMM force field predicts a helix formed from residues 5 to 15, while AMBER predicts a helix formed from residues 3-12 (Fig. S8C, Fig. S9C). The overall helicity is consistent among all simulations started with all three initial structures (2LD0, 2LD2 and Helix), even though the detailed percentage can differ somewhat from one simulation to another. In short, all three force fields yielded high helicity values (50∼70%) which agree well with the experimental values (calculated with crystal structure 2LD2). Therefore, our results of probing the helicity of HTT-N17 on micelles are comparable with the experimental data obtained under similar conditions (with the same lipid and same salt concentration)10.

The helicity of HTT-N17 is significantly reduced in water In amphipathic environments, such as micelles and TFE solution5, or crystallization conditions facilitated by fusing proteins, the HTT-N17 region adopts predominantly helical structures11-13. However, the HTT-N17 protein is also believed to be largely unstructured in aqueous solution10. Representative structures of HTT-N17 in aqueous solution from all three starting structures (Fig. 1A, Fig. 1B and Fig. 1C) are shown in Fig. 1D, Fig. 1E and Fig. 1F. Here, both OPLS-AA (D) and CHARMM (E) force fields yielded a random coil feature, resulting in a

Molecular Mechanism of Stabilizing the Helical ... - ACS Publications

Recommend Documents