Mechanism of Trehalose-Induced Protein Stabilization from Neutron

ACS2GO © 2019. ← → → ←. loading. To add this web app to the home screen open the browser option menu and tap on Add to homescreen...
0 downloads 0 Views 657KB Size
Subscriber access provided by UNIV OF LOUISIANA

B: Biophysics; Physical Chemistry of Biological Systems and Biomolecules

Mechanism of Trehalose Induced Protein Stabilization from Neutron Scattering and Modeling Christoffer Olsson, Samuel Genheden, Victoria García-Sakai, and Jan Swenson J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.9b01856 • Publication Date (Web): 09 Apr 2019 Downloaded from http://pubs.acs.org on April 11, 2019

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

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

Page 1 of 38 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

Mechanism of Trehalose Induced Protein Stabilization from Neutron Scattering and Modeling

Christoffer Olsson*[a], Samuel Genheden[b], Victoria García Sakai [c], Jan Swenson[a]

[a] Department of Physics, Chalmers University of Technology, SE-412 96 Göteborg, Sweden E-mail: [email protected] [b] Deparment of Chemistry and Molecular biology, University of Gothenburg, Box 462, SE-405 30 Göteborg, Sweden [c] ISIS Facility, STFC Rutherford Appleton Laboratory, Harwell Campus, Didcot, OX11 0QX, Oxfordshire, United Kingdom Corresponding Author: Christoffer Olsson *E-mail: [email protected]

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 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 38

ACS Paragon Plus Environment

2

Page 3 of 38 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

Abstract: The sugar molecule trehalose has been proven to be an excellent stabilizing cosolute for the preservation of biological materials. However, the stabilizing mechanism of trehalose has been much debated during the previous decades and it is still not fully understood, partly because it has not been completely established how trehalose molecules structure around proteins. Here we present a molecular model of a proteinwater-trehalose system, based on neutron scattering results obtained from neutron diffraction, quasielastic neutron scattering and different computer modeling techniques. The structural data clearly show how the proteins are preferentially hydrated, and analysis of the dynamical properties show that the protein residues are slowed down because of reduced dynamics of the protein hydration shell, rather than because of direct trehaloseprotein interactions. These findings are thereby giving strong support for previous models related to the preferential hydration model and contradicting other models based on water replacement at the protein surface. Furthermore, the results are important for understanding the specific role of trehalose in biological stabilization, and more generally for providing a likely mechanism of how co-solutes affect the dynamics of proteins.

ACS Paragon Plus Environment

3

The Journal of Physical Chemistry 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 38

Introduction Most biological molecular activities occur in the presence of water. Proteins in a cell are hydrated by a layer of water, and thus it is not surprising that the structure and dynamics of proteins are often interwoven with the structure and dynamics of water. Many studies have focused on investigating the coupling between water and protein motions; these address different aspects, such as what modes of motion of the protein are enabled by the water dynamics, and what is the precise nature of these mechanisms.1–8 There are many important reasons for exploring these questions; one important topic is the stabilization of complex biological molecules, such as proteins. By gaining deeper insight into how proteins and other biological molecules are stabilized, we can improve preservation techniques and protocols, essential for the storage of pharmaceuticals or human organs prior to transplants. Critical to most preservation techniques is the addition of stabilizing co-solvents. A good stabilizing co-solvent should ideally reduce the motions of the biological structures, thus preserving them in their native states, such that when the structures are thawed or rehydrated, they retain their original functions. The disaccharide trehalose has been shown to be a good stabilizing co-solute for several preservation techniques.9–12 For example, trehalose has been shown to protect biological materials from heat-shock13–17 and desiccation18,19. For this reason, we choose trehalose as a co-solute for this study and explore the mechanism behind protein stabilization. Understanding the specific stabilization mechanism of trehalose may provide needed insight into how proteins are stabilized more generally. One important property of trehalose is its glass-forming capability. Trehalose has a high glass transition temperature, Tg, and a high viscosity, compared to other similar compounds.20 At first, this aspect was proposed as the main property behind protein stabilization; slower dynamics in the

ACS Paragon Plus Environment

4

Page 5 of 38 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

solution implies slower motions of the proteins, which, in turn, leads to stabilization.20 It was later shown that simply using a glass former with a higher Tg did not necessarily improve protein stabilization.21–23 In fact, some of these glass formers provided negligible stabilization effects, pointing towards more direct interactions between the stabilizing co-solute and the proteins, as a necessary condition to obtain a good stabilizing environment. Various newer models have been proposed, such as the water replacement model24 and the preferential hydration model.25–28 According to the water replacement hypothesis, the protein dynamics are slowed down due to faster water molecules at the protein surface being replaced with slower moving trehalose molecules, which subsequently stabilizes the proteins. According to the preferential hydration model, however, trehalose is preferentially excluded from the protein surface, thus retaining the protein’s native hydration shell, although with a dynamic coupling between the protein and the trehalose via the hydration layer.29,30 This latter scenario has been suggested from molecular dynamics (MD) simulations of almost dried protein/trehalose systems, which have indicated that a layer of water is trapped at the protein surface,31 and that the protein dynamics become reduced due to a relatively strong dynamical coupling between the protein and the trehalose via the water layer.29,32–36 Several studies have also shown that trehalose slows down the dynamics of water, more than other similar disaccharides37–41, which means that the bioprotective effect of trehalose can be related to its greater ability to couple its dynamics to a protein with a relatively strong reduction of the dynamics in the hydration layer. In more diluted systems, on the other hand, MD simulations42,43 suggest that trehalose forms structures outside the first water layer of the protein. This implies that the protective effect of trehalose in more diluted systems is similar to that in the almost dry state. Most of these conclusions come from results from MD simulations and indirect

ACS Paragon Plus Environment

5

The Journal of Physical Chemistry 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 6 of 38

experimental measurements, such as vibrational spectroscopy methods30,44, but have not been validated by structural experimental data. In the present paper we elucidate these structural and dynamical properties by direct experimental investigations and complementary modeling. The structural picture obtained clearly shows that trehalose molecules are preferentially excluded from directly interacting with the protein surface, in support of the preferential hydration model. This fitted well with the results from QENS which showed that the dynamics of both protein and water shifted significantly with the addition of trehalose, however, the trehalose dynamics was relatively unperturbed by the addition of the protein. This agrees with a picture where trehalose molecules are not directly interacting with the slow dynamics of the protein. Rather, it suggests that the protein dynamics is reduced through an indirect coupling between the protein and the trehalose mediated via the water layer. This last indication was further validated by MD simulations.

Experimental Section

Sample preparation

Six different samples for the three-component (water–trehalose–myoglobin) system, with the same molar concentration of 1956:51:1 (corresponding to a weight ratio of 2:1:1 for the fully protonated sample) were prepared. These six samples differed however in isotope composition accordingly: 1. D2O - d-trehalose- Myoglobin (with deuteration of all exchangeable OH-groups) 2. D2O - h-trehalose - Myoglobin (with deuteration of all exchangeable OH-groups) 3. H2O - d-trehalose - Myoglobin 4. H2O - h-trehalose - Myoglobin

ACS Paragon Plus Environment

6

Page 7 of 38 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

5. 50-50 mol% mixture of samples 1 and 2 6. 50-50 mol% mixture of samples 1 and 4.

Three samples were also made with only water (H2O, D2O, and HDO) and myoglobin (1956:1 water:myoglobin molar ratio), as a reference. Protonated α,α-trehalose (or D-(+)-trehalose), referred to as h-trehalose, or h-Tre from now, was purchased from Sigma-Aldrich in dihydrate form, and used without any further purification in sample 4. The deuterated α,α-trehalose (referred to as d-trehalose, or d-Tre from now) was purchased from Omicron Biochemicals in anhydrous form. In this sugar, the carbon-bound hydrogens had been deuterated. The hydroxyl groups in sample 1 and sample 2, were deuterated by dissolving these sugars in D2O followed by drying the solutions under vacuum at 70°C. This procedure was repeated several times to ensure that all hydroxyl groups became deuterated. Myoglobin was also purchased from Sigma-Aldrich and was freeze-dried in either H2O (for samples 3 and 4) or D2O (for samples 1 and 2) before use. This was done to remove residual water molecules, and to deuterate the exchangeable protein hydrogens. The pH of the different solutions was determined to be approximately 7.8 for fully deuterated samples and 6.0 for fully protonated samples.

Neutron diffraction experiments The samples were transferred to standard 1 mm thick Ti0.676Zr0.324 containers, which were sealed with a PTFE O-ring. These were then mounted inside the neutron diffractometer on an automatic sample changer connected to a water bath for temperature control (300 K). The neutron diffraction experiments were performed on the Near and InterMediate Range Order Diffractometer (NIMROD)45 at the ISIS Pulsed Neutron and Muon Source (STFC Rutherford Appleton

ACS Paragon Plus Environment

7

The Journal of Physical Chemistry 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 8 of 38

Laboratory, U.K.). The experimentally measured raw data were corrected using the GUDRUN suite (2015 version)46. More details on the analysis of the small-angle neutron scattering, and the form factor calculations can be found in the SI.

EPSR modeling For the intermediate to small-scale structures, the corrected diffraction data in the range 0.3 – 50 Å-1, was modeled using the Empirical Potential Structure Refinement (EPSR) method.47 EPSR is a Monte Carlo based method in which an atomistic model of the investigated system is built and assigned with a reference potential. An empirical potential – based on the difference between the simulated and experimental structure factors (S(Q)) is then added to the reference potential, followed by an equilibration of the system based on the new potential. This process is then iterated until the difference between the simulated and experimental structure factors becomes negligible. More details on this method can be found in e.g. Ref. 47. In the present EPSR simulation, we produced a cubic simulation box (with a volume of 46x46x46 Å3) containing one myoglobin molecule, 51 trehalose molecules, and 1956 water molecules, which corresponds to the molar concentration of the experimental samples. The number density of the three-component samples was determined to be 0.1094 atoms Å-3. The reference potential was set identical to that in Ref. 48 and Ref. 49 for the water (from the SPC/E force field) and trehalose molecules (from OPLSAA force field50), for which the Lennard-Jones parameters are as listed in table S1. The atoms in trehalose were divided into the following different categories: Ot atoms are the oxygens in the hydroxyl groups, O1 is the glycosidic linkage oxygen, O2 are the oxygens in the aromatic rings, and O3 are the oxygens belonging to the hydroxymethyl groups. The protein structure was obtained from the protein data bank (PDB ID: 1DWR51 ). The reference potential for this protein

ACS Paragon Plus Environment

8

Page 9 of 38 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

was loosely based on the OPLSA-AA force-field52,53. All atoms of a specific element were grouped together as the same type, e.g. a hydrogen atom bonded to a nitrogen atom was grouped in the same category as a hydrogen bonded to an oxygen atom. The Lennard-Jones and Coulomb parameters for the different atoms are also presented in table S1. The EPSR simulation ran until an increase of the empirical potential no longer had an effect on the goodness of the fits of the simulated structure factors to the experimental ones. After such refined atomistic model had been obtained, the program produced 3000 configurations of the system to gather statistics of the acquired pair-correlation functions. The resulting fits and the experimental data are shown in figure 1 (diffraction data of trehalose in water can be found in Ref. 49). The relatively good agreement with all the experimental data sets implies that the structural model produced by EPSR can be regarded as a representative model of the local structure around each protein molecule. The resulting model was also validated by a) running the simulation with a new (different from the first simulation run) randomized distribution of the solutes and b) starting the simulation with trehalose molecules preferentially bonded to the protein surface (by introducing an initial artificial protein–trehalose attractive force, which was later turned off). In both cases, the final results reproduced similar fits and partial correlation functions. The analysis of the model was partly made by creating a “pseudo-trajectory” using 1000 configurations from the EPSR simulation as “molecular dynamics frames”. This trajectory could then be analyzed using common molecular dynamics trajectory analysis software, such as visual molecular dynamics (VMD)54. QENS measurements QENS measurements were performed on the IRIS spectrometer at the neutron spallation source ISIS (Rutherford Appleton Laboratory, U.K.). More detailed information about this instrument can

ACS Paragon Plus Environment

9

The Journal of Physical Chemistry 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 10 of 38

be found in Ref.55. Measurements were carried out at a temperature of 300 K on the following isotope compositions: Mb in d-Tre and H2O, Mb in h-Tre and H2O, Mb in h-Tre and D2O, Mb in D2O, Mb in H2O, h-Tre in D2O, and d-Tre in H2O. The samples were placed in annular aluminum alloy cans with a sample thickness of 0.1 or 0.25mm (depending on the hydrogen content in the sample, to ensure a sample scattering of less than 10%). The PG002 analyzer configuration was used, which gave an experimental energy transfer window of ±0.4 meV and an energy resolution of 17.5µeV FWHM. The scattered signal was collected on 51 detectors spread over different scattering angles, which were grouped into 10 groups corresponding to a Q-range of 0.42 – 1.8 Å-1. The measured spectra are a convolution of the actual dynamic structure factor of the sample (S(Q,)), and the resolution function (R(Q,)). R(Q,) was measured by taking the spectra of a vanadium standard, and thus S(Q,) was obtained by de-convoluting the sample spectra.56 The Fourier transform of the dynamic structure factors, S(Q,), denoted as the intermediate scattering functions (I(Q,t)), were used for the analysis in this study. The I(Q,t):s were cut at 150 ps since the data are not reliable at longer time-scales due to the limited resolution of the spectrometer. The Mantid software package56 was used for all the data corrections and analysis (e.g. Fourier transformation and deconvolution).

QENS Fitting procedure The I(Q,t):s were fitted by assuming a single stretched relaxation for each of the species in the samples, with a relaxation time of  and a stretching parameter  (where 0