Reorientation Motion and Preferential Interactions ... - ACS Publications

Subscriber access provided by MAHIDOL UNIVERSITY (UniNet)

Article

Reorientation Motion and Preferential Interactions of a Peptide in Denaturants and Osmolyte. Gouri S. Jas, Eric C Rentchler, Agnieszka S#owicka, John R. Hermansen, Carey K Johnson, C. Russell Middaugh, and Krzysztof Kuczera J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.6b00028 • Publication Date (Web): 11 Mar 2016 Downloaded from http://pubs.acs.org on March 16, 2016

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 35

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

Reorientation Motion and Preferential Interactions of a Peptide in Denaturants and Osmolyte.

Gouri S. Jas1*, Eric C. Rentchler2, Agnieszka M. Słowicka3, John R. Hermansen4, Carey K. Johnson2, C. Russell Middaugh1, Krzysztof Kuczera2,5 1

Department of Pharmaceutical Chemistry, The University of Kansas, Lawrence, KS 66047 2

3

Department of Chemistry, The University of Kansas, Lawrence, KS 66045

Institute of Fundamental Technological Research, Polish Academy of Sciences, ul. Pawińskiego 5B, 02-106 Warsaw, Poland 4 5

Central University of the Caribbean, School of Medicine, Bayamon, PR 00956

Department of Molecular Biosciences, The University of Kansas, Lawrence, KS 66045

Corresponding Author: Gouri S. Jas Department of Pharmaceutical Chemistry The University of Kansas, Lawrence, KS 66047 [email protected]

ACS Paragon Plus Environment


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. Fluorescence anisotropy decay measurements and all atom molecular dynamics simulations are used to characterize the orientational motion and preferential interaction of a peptide, N-acetyl-tryptophan-amide (NATA) containing two peptide bonds, in aqueous, urea, guanidinium chloride (GdmCl), and proline solution. Anisotropy decay measurements as a function of temperature and concentration showed moderate slowing down of reorientations in urea and GdmCl and very strong slowing down in proline solution, relative to water. These effects deviate significantly from simple proportionality of peptide tumbling time to solvent viscosity, leading to the investigation of microscopic preferential interaction behavior through molecular dynamics simulations. Examination of the interactions of denaturants and osmolyte with the peptide backbone uncovers the presence of strongest interaction with urea, intermediate with proline and weakest with GdmCl. In contrast, the strongest preferential solvation of the peptide sidechain is by the nonpolar part of the proline zwitterion, followed by urea, and GdmCl. Interestingly, the local density of urea around the sidechain is higher, but the GdmCl distribution is more organized. Thus, the computed preferential solvation of the sidechain by the denaturants and osmolyte can account for the trend in reorientation rates. Analysis of water structure and its dynamics uncovered underlying differences between urea, GdmCl and proline. Urea exerted the smallest perturbation of water behavior. GdmCl had a larger effect on water, slowing down kinetics and stabilizing interactions. Proline had the largest overall interactions, exhibiting a strong stabilizing effect on both water-water and water-peptide hydrogen bonds. The results for this elementary peptide system demonstrate significant differences in microscopic behavior of the examined solvent environments. For the commonly used denaturants, urea tends to form disorganized local aggregates around the peptide groups and has little influence on water, while GdmCl only forms specific interactions with the sidechain and tends to destabilize water structure. The protective osmolyte proline has the strongest and most specific interactions with the tryptophan sidechain, and also stabilizes both water-water and water peptide hydrogen bonds. Our results strongly suggest protein or peptide denaturation triggered by urea occurs by direct interaction, whereas GdmCl interacts favorably with sidechains and destabilizes peptide-water hydrogen bonds. The stabilization of biopolymers by an osmolyte such as proline is governed by favorable preferential interaction with the sidechains and stabilization of water.


Page 2 of 35

Page 3 of 35

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. Co-solvents are known to modify protein-protein interaction and are extensively used to gain insight into peptide and protein folding and aggregation1-5. Small organic molecules such as urea and guanidinium chloride (GdmCl), chaotropes, destabilize ordered protein structures whereas many polyols, sugars and free amino acids stabilize the native form

6-11

.

These molecules are known as cosolvents because they occupy a significant fraction of the solution volume12. Even though there is a long history of the use of denaturants and osmolytes to manipulate the structural integrity of biologically active macromolecular systems, the microscopic details of the underlying interactions between co-solvents and a native or denatured proteins are yet to be fully understood. Preferential interactions are believed to play an important role in co-solvent induced protein destabilization and stabilization as well as cellular stress. In order to obtain a more detailed description for the preferential interactions of co-solvents with peptides, Kirkwood-Buff type integrals have been used to picture the co-solvent distribution around a protein. Another important component of examining the effect of co-solvents on biomolecules is to investigate the nature of water structure in the environment of a co-solvent. GdmCl (a salt) and urea (a polar molecule) can modulate the strength of hydrophobic interactions, based on the degree to which the local water structure is disrupted13. Chaotropes are thought to disrupt water structure and kosmotropes to order water and subsequently increasing and decreasing the strength of hydrophobic interactions, respectively. These compounds can also perturb salt bridge formation in proteins. Factors that affect these interactions are critical to our understanding the molecular origin of the Hofmeister series13,14 in which ions are ranked based on their ability to salt in and out proteins. To identify specific mechanisms and to better understand interactions between small organic molecules like urea and gunidinium chloride and polypeptide chains experiments and simulations of model systems and proteins have been employed. Simpler model systems have been designed to probe the effect of charge and polarity and the role of water. Conflicting results concerning a reduction in the hydrophobic effect as a result of the presence of urea have arisen from studies of small hydrophobic solutes in urea solutions15-19. In molecular dynamics simulations, however, a larger effect of urea has been observed in model hydrophobic systems. A purely hydrophobic polymer was shown to unfold in urea due to enthalpically favorable dispersion interactions20. It has also been shown that urea tends to aggregate around and in the interior of carbon nanotubes21,22. Therefore, energetically favorable interaction with purely



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

hydrophobic groups can modulate the hydrophobic effect in urea. The presence of favorable interaction of urea was also shown for charged solutes, involving hydrogen bonding and disruption of ion pairs15,19. In the case of GdmCl, it has been suggested that interaction between the positively charged guanidinium group (Gdm+) and charged residues as well as the peptide backbone is the dominant mechanism of proteins destabilization in aqueous GdmCl solution19. It has also been suggested that the existence of a strong direct interaction of Gdm+ with charged species19 compared to that of urea may explain the enhanced efficiency of denaturation of proteins by GdmCl compared to urea. Another path of protein denaturation is termed the indirect mechanism, in which urea causes the perturbation of protein structure by the alteration of water structure. This view surfaced as a result of transfer experiments in which hydrocarbons tend to be more soluble in aqueous urea than water23. The suggested picture from such transfer experiments was that urea alters water structure, thus weakening the hydrophobic effect by influencing the rapid solvation of alkanes24. Further examination employing both experimental and computational studies, however, demonstrated that urea integrates well into the hydrogen-bonding network of water and that water’s spatial distribution remains relatively unaffected with a minimal tendency of selfaggregation25-31. In contrast, the molecular mechanism by which the naturally occurring osmolyte proline stabilizes proteins is not clear. It has been suggested that proline preferentially interacts with the sidechains but unfavorably with the backbone and that the sum of these effects may be a reason why proline may rescue aggregation processes during folding events 32. Studies of molecular reorientations in solution provide the most direct access to the rates of nanosecond and sub-nanosecond molecular motions and interactions with the environment. The interpretation of the dynamics of a simple peptide such as the N-acetyl-tryptophan-amide (NATA)33-35 can be very useful for describing the dynamic nature of larger systems such as helices, hairpins, and proteins. An important component of this study is to obtain baseline measured data in the time domain that can be extensively modeled with all atom molecular dynamic simulations under a variety of conditions to describe the molecular origin of the observed behavior. To explore reorientational motion which directly probe interactions of a peptide with denaturants and osmolytes, we have carried out fluorescence anisotropy decay measurements of NATA in


Page 4 of 35

Page 5 of 35

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


aqueous, urea, GdmCl, and proline solution in temperatures from 280-320K at a solute concentration from 0-5.5 M. All-atom molecular dynamics simulations of NATA in aqueous, 5 M urea, 5 M GdmCl, and 5 M proline solution at 280K, 300K, and 320K are performed to identify the microscopic mechanism of interactions and their influence on this elementary peptide system. Measured peptide orientation times were found to be slowest in proline solution followed by GdmCl, and urea. Significantly slower global orientation times indicate an underlying preferential interaction of the peptide with its surroundings. Analysis of hydrogen bond kinetics show that the rate of water-water hydrogen bond breaking is significantly slower compared to water in all systems except urea. For peptide systems, our results indicate that the most commonly used chaotrope, urea, engages in the strongest preferential interactions with the backbone, proline interacts most strongly with the indole sidechain whereas the weakest interaction are present with GdmCl. Results and Discussion. Fluorescence anisotropy decay. The osmolyte and denaturant induced motion of a simple peptide, NATA (with two peptide bonds, figure 1), and its direct interaction with the environment is examined with fluorescence anisotropy measurements. Figure 1. Shown here is the chemical structure (A) and ball-and-stick representation (B) of NATA with two peptide bonds. In A, blue and red arrow represent the rotation around φ and ψ, respectively.

We have used two chaotropes, urea and guanidinium chloride, and the osmolyte proline, with concentrations up to 5.5M over a temperature range of 280-320K. At 298K the solvent viscosities are ~1.3 cp in 5M urea36, ~1.5 cp in 5M GdmCl36, ~7 cp in 5M proline37, and ~1cp in water.

A summary of the fluorescence lifetime and global orientation time of NATA is

presented in figure 2 (and as a table in the supplementary information). Representative fluorescence intensities in parallel and perpendicular polarization and anisotropy decay at 320K in four different solvent conditions are shown in figure 3(A-D). Fluorescence of NATA is observed to be substantially quenched in the presence of proline co-solvent possibly due to proton and/or electron transfer38,39. This does not influence the anisotropy that is determined from the measured fluorescence intensity ratio. Rotational orientation times (τrot) determined from fluorescence anisotropy measurements are slower by a factor of 2.4 in 5.5M urea, 3.3 in



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

5.5M GdmCl, and 10.3 in 5.5M proline compared to water at 298 K. Viscosities of 5.5M urea, GdmCl and proline are higher by factors of about 1.3, 1.5 and 7, respectively, with respect to water at 298K. Although viscosities of these co-solvents are higher than water, a simple viscosity effect cannot account for the significantly slower orientation time in these co-solvent environments. One of the most important observations concerning these anisotropy measurements is that the global orientation time of the peptide in 5.5M proline is slower by a factor of 10.3. In urea and GdmCl, the co-solvent viscosity is slightly higher than water but the rotational orientation time is slower by a factor of 2.4 and 3.3, respectively. At 298K, a twoexponential fit to the anisotropy decay extracted a fast component (φ1) with a time constant around 11 to 21 ps and a slow component (φ2) with a time constant of 30 to 304 ps in 0 - 5.5M denaturant and osmolyte concentration. The fast rotational correlation time corresponds to internal dynamics involving the sidechain indole ring. The slow rotational correlation time measures the global tumbling, and thus the peptide’s interaction with its immediate environment. Figure 2. Summary of temperature and concentration dependent measured fluorescence lifetime (A-C) and rotational orientation (φ2) time (D-F) of NATA in denaturants and an osmolyte. Measured in four different concentrations (0 to 5.5 M) and three different temperatures (280K, 300K, and 320K) for urea and GdmCl and three different concentrations (0M, 3.5M, and 5.5 M) and three different temperatures (280K, 300K, and 320K) for proline.

Since the effects of solvent viscosity alone were unable to account for the significant slowdown in the measured global orientation time, this led us to investigate the environmental effects in the form of preferential interactions of the co-solvents with the peptide. The experimental observations indicate that the peptide in proline solution interacts with the environment very differently (φ2 = 304 ps in 5.5M at 298K) compared to urea (φ2 = 71 ps in 5.5M at 298K) and GdmCl (φ2 = 98 ps in 5.5M at 298K) relative to water (φ2 = 30 ps). Study of the radial distribution, preferential interaction, hydrogen bond kinetics, and water relaxation with all atom molecular dynamics in these co-solvents independently rationalize these experimental observations. Figure 3. Time-correlated single-photon counting is employed to measure fluorescence decays for NATA in proline, urea, and GdmCl solution. Fluorescence decay polarized parallel to the excitation (blue), perpendicular to the excitation (green), and the instrument function (black) 0M (A) and 5.5M (proline B, Urea C, and GdmCl, D) at 320K. The decays for fluorescence with polarization perpendicular to the excitation polarization are about 30% of the intensity of fluorescence with parallel polarization due to the different detection efficiencies in the two detection


Page 6 of 35

Page 7 of 35

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


channels (see Methods). Fit to the parallel and perpendicular fluorescence decays are in red. Weighted residuals are for the fits to the parallel (blue) and perpendicular (red) fluorescence decays. The fluorescence anisotropy is presented in the inset. For comparison, the C2(t) correlation function from explicit cosolvent molecular dynamics simulations of NATA illustrating rotational orientation time in 0M and 5M proline, urea and GdmCl at 300K and 320K are presented in figure 3E and 3F. Orientation times from fluorescence anisotropy measurements and molecular dynamics simulation in 0M and 5M proline, urea and GdmCl at 300K are shown in figure 3G.

Molecular Dynamics Simulations in Aqueous, Denaturants, and Osmolyte Solution. To address the relation between the observed dynamics of the peptide in denaturants and osmolyte and the atomic behavior responsible for these observations, we have carried out all atom molecular dynamics simulation in the same co-solvents used in the experimental studies.

Rotational diffusion. At 300 K the rotational orientation time (τrot) from molecular dynamics simulations are determined to be 30 ps in water, 44 ps in urea, 104 ps in GdmCl and 340 ps in proline, using the OPLS/AA force field for NATA and the TIP4P water model (figure 3G). Figure 4. Measures of water dynamics from MD simulations at 300 K. Translational diffusion coefficients for water (A) and NATA (B), water reorientation time from dipole autocorrelation function C1(t) (C) τ1d, ps. Rates (D,E) and activation energies (F,G) of hydrogen bond breaking (forward) and formation (backward) processes and relaxation times (H) for water-water and water peptide hydrogen bonds in co-solvent environments.

These results are in good agreement with the measured orientation (φ2) times of 30 ps in water, 71 ps in urea, 98 ps in GdmCl, and 304 ps in proline solution. C2(t) auto-correlation functions at 300 and 320 K, and a summary of peptide orientation times comparing measurements with simulation results are presented in figure 3E-G. The influence of co-solvents caused a slowdown of the peptide reorientations in urea and GdmCl and a substantial slowdown in the presence of proline. Compared to the water trajectory, rotational diffusion is slower by a factor of 1.5 in urea, 3.5 in GdmCl and 16 in proline. For GdmCl and proline, the effects are much stronger than the trend in the experimental solvent viscosities, as found in the experimental measurements. At higher temperature, (320K), the simulation correctly determined the effect of temperature which showed an increase in reorientation rates in all four environments. The correlation times (τ1d) for reorientation of the water dipoles at 300 K are shown in Fig. 4C. These data follow similar trends in peptide reorientations – motions are fastest in water (τ1d = 3.1



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

ps), somewhat slower in urea (τ1d = 3.2 ps), markedly slower in GdmCl (τ1d = 8.0 ps) and slowest of all in proline (τ1d = 12 ps). Here the effect of GdmCl is stronger, while the effect of proline is weaker than predicted by bulk viscosity changes alone. Translational diffusion. The calculated values of translational diffusion coefficients Dt for water and NATA are presented in figure 4 (A, B). For diffusion of the water component, Dt values were 3.54, 3.14, 1.65 and 1.10×10-9 m2/s, for water, urea, GdmCl and proline simulations at 300 K, respectively. The computed value of the aqueous trajectory is close to the reported selfdiffusion values for pure TIP4P water of 3.69×10-9 m2 s-1 at 298 K40. The simulations predict the slowing down of translational diffusion upon addition of co-solvents. Relative to the water trajectory, TIP4P diffusion is slower by a factor of 1.1 in urea, 2.1 in GdmCl and 3.2 in proline. The peptide Dt values at 300 K were 0.80, 0.61, 0.28 and 0.11 ×10-9 m2/s, in water, urea, GdmCl and proline, respectively. The variation of translational diffusion rates of both peptide and water with co-solvent was closer to the trend of measured solvent viscosities36,37at 300 K than found for reorientation rates. Hydrogen bond dynamics. Rate constants for hydrogen bond breaking (forward, k) and formation (backward, k’) following the formalism of Chandler and Luzar41 are presented in figure 4 (D, E), broken down into data for water-water and water-peptide hydrogen bonds. The addition of urea has a generally small effect on hydrogen bond kinetics. GdmCl induces a slowing of both water-water and waterpeptide hydrogen bond breaking (by about 40%), while speeding up water-water hydrogen bond formation by about 30%. Proline has the strongest effect on the kinetics: water-water hydrogen bond breaking is slowed down by a factor of two, while the formation rate decreases only slightly, while water-peptide hydrogen bond breaking is slowed down by a factor of 4.4, and water-peptide hydrogen bond formation by a factor of two. These results show a clear difference in mechanism of action for all three co-solvents. Urea essentially does not change the hydrogen bond dynamics. GdmCl slows down hydrogen bond kinetics, slightly strengthening water-water and moderately strengthens water-peptide interactions. Proline stabilizes both water-water and water-peptide hydrogen bonds, with the latter effect being the strongest of all examined.


Page 8 of 35

Page 9 of 35

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


Generalized diffusion. Potentials of mean force V(x) and coordinate-dependent diffusion coefficients D(x) calculated according to Bicout and Szabo42 are presented in figure 5(C, F). These data show that the time evolution of the internal structure of the peptide is generally slowed by the addition of co-solvents. Figure 5. Probability distributions, potential of mean force (PMF), and generalized diffusion coefficients of NATA in cosolvents from MD simulation. Rg - Radius of gyration and phi – dihedral angle defined in figure 1A.

Figure 6. Radial distribution (RDF) and preferential interaction of dipeptide in co-solvents. RDFs of backbone C=O of the NATA with respect to solvent nitrogen of urea, GdmCl, proline in 6A. RDFs of backbone nitrogen of the peptide with respect to the solvent nitrogen of urea, GdmCl, proline in 5B. In 5C, the sidechain, tryptophan relative to nitrogen of urea, GdmCl, and proline. Shown in 5D, the RDF of tryptophan relative to proline Cα, Cβ, Cγ, Cδ and sum of proline sidechain carbons. Backbone carbonyl (C=O), backbone N- nitrogen of backbone, Nproline – nitrogen of Proline, Nurea – nitrogen of urea, NGdmCl – nitrogen of GdmCl, W- water; Cg proline – gamma carbon of proline, Ca proline – alpha carbon of proline, Cb proline – beta carbon of proline, Cd – delta carbon of proline, Ourea – oxygen of urea, Owat – oxygen of water.

The effect of co-solvent on internal motions, however, is much weaker than the strong influence found on translational and rotational diffusion. Typically, values of D(x) in the presence of cosolvents are lower than those in water by not more than 20%. This indicates that for NATA the internal friction effect is not as strong as external (solvent) friction.

Solvation. The microscopic effects underlying differences of peptide dynamics in the presence of cosolvents were further studied by analyzing solvent structure. Results for radial distribution functions (RDF) are shown in Fig 6 and preferential solvation effects in Fig. 7 for 300 K MD trajectories. For urea strong preferential peptide solvation is observed (Figs. 6A-C). Individual urea molecules interact directly with the peptide backbone (figure. 6AB) and sidechain (figure 6C). The RDFs exhibit both a strong first solvation peak and a significant, broad second peak at 5-10 Å. Within a 10 Å Figure 7. Traces of preferential interaction and preferential coordination numbers are presented. Figures 7 A, B, C show the preferential interaction of backbone nitrogen and backbone carbonyl oxygen relative to nitrogen and oxygen of urea and proline and nitrogen of GdmCl. Figure 7 D, E, present the preferential coordination number of urea, proline, and GdmCl relative to backbone and sidechain. In figure 7F, is shown the preferential coordination of the solvent structure, oxygen of water with respect to nitrogen and oxygen of urea and proline, and nitrogen of



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 35

GdmCl. bbN – backbone nitrogen, bbO –backbone carbonyl oxygen, NGdm- guanidinium nitrogen, ou - urea oxygen, nu - urea nitrogen, Nprol – proline nitrogen, Oprl – proline oxygen.

sphere, there is a local excess of 4 ureas around the peptide, preferentially solvating both the backbone and the sidechain, relative to the number expected from average densities (Figure 6). In the case of GdmCl, we see an essentially featureless distribution, with the guanidinium cation mostly depleted compared to average density up to 9 Å from NATA (figure 6AB). Consequently, the density of guanidinium within 10 Å of the peptide is essentially the same as the average density (figure 7). In the proline system the situation is intermediate between the other two cosolvents. Proline exhibits a weak first solvation peak and a broad second solvation shell in the 510 Å range, which is not as pronounced as in urea (figure 6 AB). In terms of preferential solvation, we find 2 excess Proline zwitterions within a 10 Å sphere of the peptide backbone (figure 7). The indole sidechain solvation is markedly different than that of the backbone. For the indole ring, we find the strongest preferential solvation by the hydrophobic atoms of proline (Cβ and Cγ) which exhibit very high peaks in the 3-6 Å range, and a much lower and narrower peak for GdmCl (figure 6 C, D). The sidechain solvation by urea is intermediate between proline and GdmCl, exhibiting a very broad, unstructured distribution in the 3-10 Å range (figure 6C). Distributions of water oxygens (OW) around other water oxygens showed that there were 4.7 neighbors in the first solvation shell (3.4 Å) in water, 4.2 neighbors in urea, 3.7 in GdmCl and 4.1 in proline simulations. The larger decrease in GdmCl probably reflects the fact that at equimolar concentrations, the GdmCl solutions have twice as many co-solvent molecules (Gdm+ and Cl-) as the other systems. In the GdmCl molecular dynamics simulations, the distribution of the chloride ions (Cl−) around the peptide backbone is similar to that of guanidinium (Gdm+) – it is mostly featureless, representing close to average density throughout the system. There is no evidence of preferential solvation by chloride around the backbone or sidechain of NATA (Graphical representations are presented in the supplementary information). Figure 8. Average density maps and snapshots of co-solvent molecules relative to the NATA. (A,B) in urea; (C,D) in GdmCl; (E,F) in Proline. The density maps are 1 A thick slices in xz plane, oriented so that NATA sidechain is in xy plane with long axis along x. Light blue corresponds to lowest and white to highest density.


Page 11 of 35

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


Sample snapshots from the trajectories showing positions of the co-solvents and co-solvent density around the NATA sidechain are illustrated in figure 8. These images confirm our previous analysis. Urea has a strong tendency for interactions with the peptide, but its solvation shell is loosely organized. Proline has a smaller number of molecules interacting with the peptide, but they are highly localized below and above the indole ring. Graphical analysis shows some interesting highly selective interactions, in which the charged groups of Pro interact with the peptide blocking groups while the nonpolar atoms of proline simultaneously contact the indole ring. GdmCl structures are intermediate in nature – they do not approach the peptide as closely as urea or proline, but their locations are more specific than those of urea but less so than proline. Figure 9. NATA structures from cluster centers from 300 K MD trajectories. Columns one to three show top 3 clusters by population, column four shows an example of a low population structure. Rows: (A) blue – in aqueous solution, (B) red – in 5M urea solution, (C) green - 5M GdmCl, (D) Orange - 5M proline solution.

In summary, the preferential solvation of the peptide backbone was strongest in urea, intermediate in proline and weakest in GdmCl. For the sidechain, the strongest preferential interaction was for the apolar atoms of proline, which had the most pronounced excess density over a wide range, intermediate for urea, which exhibited a very broad and mostly unstructured local density increase, and weakest for GdmCl. The well-defined specific solvation peaks found for GdmCl and proline suggest that these co-solvent molecules move together with the NATA sidechain, increasing its effective hydrodynamic size. This provides a microscopic explanation for the significant slowdown of molecular reorientations in the presence of these two co-solvents. Peptide structural preferences. Central structures for major clusters found in the 300 K NATA trajectories are shown in Figure 9. Two-dimensional distributions of the sampled backbone (φ,ψ) and sidechain (χ1,χ2) dihedrals are shown in figure 10. In terms of backbone, two main conformers were sampled, B1 at (φ,ψ) ≈ (-130o, 0o) and B2 at ≈ (-100o, -20o) with some additional population in the extended region. For sidechains there were 4 main conformers with S1 at (χ1,χ2) ≈ (60o, 90o), S2 at (-140o, -90o), S3 at (60o, -90o) and S4 at (-140o, 90o). In water the top four clusters accounted for 83% of structures



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 12 of 35

and sampled conformers (B1,S1), (B2,S2), (B1,S3) and (B2,S4), respectively. In urea the top four clusters had a total population of 84%, sampling conformers (B1,S1), (B2,S4), (B1,S3) and (B2,S2). In GdmCl the top four clusters had an aggregate population of 82% and sampled conformers (B2,S2), (B1,S3), (B2,S4) and (B1,S1). In proline the top four clusters accounted for 88% of structures and sampled conformers (B1,S1), (B1,S3), (B2,S4) and (B2,S2), respectively. Generally, the peptide explores the same conformations in the four environments, but with slightly different weights. Compared to water, the largest conformational shift is in GdmCl, which is the only environment favoring the (B2,S2) form. The sampled structures correspond primarily to extended N-terminal part of the backbone (acetyl group and N-H) and a bent Cterminal part (C=O and amide blocking group), with the indole ring pointing away from the backbone. Figure 10. Shown the (φ,ψ) and (χ1,χ2) distribution plots of NATA in the form of potential of mean force, V(x,y) = RTlnP(x,y), with R – gas constant T=300K and P- probability. Coloring: red – lowest free energy, purple – highest free energy. Backbone (φ,ψ) distribution in water (A), in urea (B), in Gdmcl (C), in Proline (D). Sidechain (χ1,χ2) distribution in water (E), in urea (F), in GdmCl (G), in Proline (H). Central structures of structural clusters corresponding to the main free energy minima are shown.

Figure 11. A color-coded map showing the interaction of the NATA with co-solvents and transparent area represents no presence of preferential interaction. (A) Red areas on the polar part of the backbone and the sidechain represent peptide interactions with urea, which are strongly preferential compared to water but not highly specific with respect to location of individual co-solvent molecules. (B) Faded green area on the peptide sidechain represents interaction with guanidine, which is mostly excluded from the peptide vicinity but is able to form specific interactions with the sidechain. (C) Orange area on the peptide sidechain represents interactions with proline, which are both preferential with respect to water and specific in location of individual prolines.

Discussion. Driven by the need to develop a detailed microscopic mechanism of the influence of denaturants and osmolytes on the structural stability of peptides and proteins, we examined the local co-solvent environment of denaturants and osmolytes and its interaction with NATA. Reorientation motion of the peptide under these cosolvents directly probed the interaction based on the type of cosolvent molecules and all atom molecular dynamics simulation visualized and constructed an atomic detail description of these interactions with respect to peptide backbone and sidechain. An important atomically detailed picture concerning the water structure and


Page 13 of 35

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


hydrogen bond kinetics in the presence and absence of denaturants and osmolyte is has been developed based on this study. Urea is known to efficiently form hydrogen bonds with water molecules as well as with peptide backbone or charged species in the absence of restrictions due to excluded volume interactions. It has been suggested that the water structure in urea remain unperturbed even in high denaturant concentration as urea efficiently form hydrogen bonds43-46. In the case of GdmCl, work by O’Brien et al. suggested that not only electrostatic interaction between Gdm+ and the charged residues but also with the peptide backbone is responsible for the destabilization of structure in aqueous GdmCl solution19. It has also been suggested that the effect of Gdm+ on water structure is ineffective in promoting dispersal of hydrophobic group in water47. The natural osmolyte proline is known to protect from aggregation of protein by suppressing anomalous protein interactions that promote aggregation32. As a mechanism of action, it has been suggested that the peptide backbone is preferentially buried in the presence of osmolytes. This provides a reasonable thermodynamic argument to explain why protein stability and folding increases with stabilizing or protecting osmolytes48,49. Based on this study we find that urea interacts strongly both with the peptide backbone and the sidechain whereas GdmCl forms a weak interaction with the sidechain alone. Based on our analysis of the water dynamics around the peptide we found that urea does not have a significant effect, whereas GdmCl slows it down. The protective osmolyte proline has the strongest and most specific interactions with the tryptophan sidechain, and also stabilizes both water-water and water-peptide hydrogen bonds. Results from this study strongly suggest peptide denaturation triggered by urea occurs by direct interaction with the backbone and the sidechain (represented with color red in figure 11 A). These interactions are found to be strongly preferential relative to water but not highly specific with respect to the location of individual co-solvent molecules. GdmCl primarily interacts favorably but weakly with sidechains (figure 11 B); it is mostly excluded from the immediate peptide surroundings but able to form specific interactions with the sidechain. The stabilization of peptides by a natural osmolyte like proline is both preferential with respect to water and specific in location of individual prolines (figure 11 C). Proline also strongly stabilizes waterwater hydrogen bonds.



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

Page 14 of 35

Conclusions. To characterize the microscopic structural basis and the presence of underlying interaction of denaturant-induced destabilization and osmolyte-triggered stabilization of biopolymers we have selected N-acetyl-tryptophan-amide (NATA), a simple peptide model system for this investigation. Anisotropy decay measurements were used to describe reorientational motions of the tryptophan sidechain in urea, GdmCl and proline solutions, in a concentration from 0-5.5 M co-solvent over the temperature range 280-320K. Measured global reorientation of the peptide slows down significantly in urea, GdmCl and proline, compared to aqueous solution. The influence of viscosity in these cosolvents alone is unable to account for the degree of increase in orientation time (τrot) with respect to aqueous solvent, leading us explore evidence of preferential interactions in these environments. All atom molecular dynamics simulations were employed to investigate the motions and the nature of specific co-solvent interactions with the peptide backbone and sidechain. The computed peptide reorientation times were quite similar to the measured data. Water properties were characterized by calculating translational and rotational diffusion, hydrogen bond kinetics and RDFs. The variation of rotational correlation times and translational diffusion coefficients for water in the simulations tended to follow the pattern of measured solution viscosities, indicating that simulations correctly reproduce bulk solvent properties. Urea has little effect on water structure or interactions. GdmCl had the strongest destabilizing effects on water structure, with fewest water-water hydrogen bonds, as well as slowing hydrogen bond kinetics. The presence of proline leads to marked slowing down of water hydrogen bond kinetics, but most markedly stabilizes water-peptide hydrogen bonding. Investigation of preferential interaction of co-solvents with the peptide backbone indicates the strongest interaction with urea, followed by proline and weakest with GdmCl. Examination of the sidechain, on the other hand, shows that the strongest preferential solvation is contributed by the nonpolar part of the proline zwitterion, followed by urea, and GdmCl. Analysis showed the local density of urea around the sidechain was higher compared to GdmCl, but GdmCl distribution is more organized. The strong specific solvation peaks of guanidinium and proline around the NATA sidechain indicate that co-solvent molecules move together with the peptide, increasing its effective size, explaining the unusually large slowing down of reorientations in these two co-solvents.


Page 15 of 35

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


Our results of this elementary peptide system demonstrate striking differences in the microscopic effects of co-solvents on biopolymers. The commonly used denaturant urea tends to form disorganized local aggregates around the entire peptide and has little influence on water. GdmCl only forms specific interactions with the sidechain and slows down water dynamics, slightly stabilizing peptide-water interactions. The protective osmolyte proline has the strongest and most specific interactions with the tryptophan sidechain, and also stabilizes both water-water and water-peptide hydrogen bonds. Our results strongly suggest peptide denaturation triggered by urea occurs by direct interaction, whereas GdmCl primarily interacts favorably but weakly, with sidechains. The structural stability of a peptide created by an osmolyte such as proline is influenced by favorable and very specific preferential interaction with the sidechains and strong stabilization of water. It would be important to test these findings in more complex systems, such as helices, hairpins and proteins, to develop a generalized mechanism. Methods. Experimental: Materials. The, N-acetyl-tryptophan-amide, was purchased from Sigms-Aldrich with the purity greater than 99%. Anisotropy measurements are conducted in 20 mM tris buffer at pH 7.2 with a 40-50µM sample concentration. Denaturant and osmolyte solutions were made in 20mM tri buffer. Fluorescence anisotropy decay measurements.

Time-correlated single-photon counting was

employed in time-resolved fluorescence measurements. NATA fluorescence is generated with, excitation at 290 nm, with the third-harmonic of a mode-locked, cavity-dumped Mira Optima 900F/Pulse Switch Ti:Sapphire laser pumped by a 10 W Verdi Laser (Coherent, Inc., Santa Clara, CA, and 5-050 Ultrafast Harmonic Generator, Inrad Northvale, NJ) with an instrument response time ≈ 10ps. NATA fluorescence was collected at 350 nm with an 8-nm bandpass (model 9030 monochromator, Sciencetech Inc, Concord, ON, Canada). As described in detail elsewhere35, parallel and perpendicular fluorescence polarizations were collected simultaneously in a T-format and processed by a PC card (Becker and Hickl, SPC-830, Berlin, Germany). Since decays of fluorescence with polarization parallel and perpendicular to the excitation polarization were collected simultaneously in two different detection channels with different detection efficiencies, a correction factor was incorporated into the fitting equations. The fitting parameter had a value g ≈ 0.3 where g is the relative detection efficiency of fluorescence with polarization



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 16 of 35

perpendicular to the excitation polarization relative to fluorescence with polarization parallel to the excitation polarization. Fluorescence decays were measured for samples in proline, urea, and GdmCl from 0 to 5.5M at 280 K, 300 K, and 320 K. Employing in-house software, fluorescence decays for parallel and perpendicular polarizations were globally fit to a 3-exponential intensity decay coupled to a double-exponential anisotropy decay. Rotational correlation times for the NATA were obtained by globally fitting the fluorescence decays with polarizations parallel and perpendicular to the excitation polarization as described elsewhere50. The fluorescence decays and their fits are shown in 2A, 2B, 2C, and 2D. In each case the anisotropy decay was fit with a two-exponential decay. The initial anisotropy r(0) at t=0 is kept fixed at 0.25 in the fitting procedure as described elsewhere51 Simulation methods. The simulated peptide system, N-acetyl-tryptophan-amide, Ac-Trp-NH2 (NATA) was studied in the indicated co-solvents. The extended structure was generated with CHARMM52, with ACE and CT2 termini. For the water simulation the peptide was solvated with 1178 TIP4P waters. For the co-solvent simulations, pre-equilibrated boxes of 5 M co-solvent with TIP4P were generated with GROMACS and used to overlay NATA. In the urea simulation, the system included NATA, 88 ureas and 950 TIP4P waters. In the GdmCl simulation, there was NATA, 84 guanidinium chloride ion pairs and 950 TIP4P waters. In the proline simulation the system consisted of NATA, 60 proline zwitterions (NH2+ and COO- termini) and 790 TIP4P waters. For each system a brief energy minimization and short simulation with restrained NATA was followed by a 1 ns equilibration at 280, 300 and 320 K under NPT conditions at 1 atm pressure. Finally, 100 ns MD trajectories were generated for each of the four systems at 280, 300 and 320 K under NVT conditions. The equilibrated box sizes for the systems were in the 31.6-33.2 Å range. The simulations used 14 Å van der Waals cutoffs and 10 Å direct cutoffs with PME for electrostatics. The GROMACS version 4.5.6 program53 was employed, using a 2 fs time step and constraints on all bond lengths, with the OPLS/AA force field54. Clustering was performed in dihedral space, using the time series of φ, ψ, χ1 and χ2 dihedrals from 300 K MD trajectories, sampled every 1 ps. The ART-2’ algorithm implemented in the CHARMM program was used 55,56

. With a 90o radius, eight clusters were identified in water, urea and proline, and seven in


Page 17 of 35

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


GdmCl. The top four clusters accounted for 83% sampled structures in water, 84% in urea, 82% in GdmCl and 88% in proline. Molecular reorientations in MD trajectories were analyzed by generating time series of transition dipole axis for the NATA side-chain 1Lb transition and calculating correlations functions C2(t) = /2, where θ is the axis reorientation during time t and the average is over starting points. The rotational correlation time τrot was obtained as the integral under the C2(t) plot. Translational diffusion was calculated from fitting mean-square displacements of water and peptide center-of-mass to a straight line in the 100-1100 ps intervals. Hydrogen bond kinetics was analyzed using the approach of Luzar and Chandler41,57, as implemented in GROMACS by van der Spoel et al.58. To describe internal dipeptide dynamics in the four environments, we have applied the treatment of Bicout and Szabo42. Briefly, evolution of selected variables x was considered to follow the discretized Smoluchowski equation, with position-dependent diffusion D(x) along a free energy surface V(x). The free energy surface, or potential of mean force (PMF) was calculated from the probability distribution histograms peq(x), as V(x) = -RT ln peq (x), with R – the gas constant, T – the absolute temperature. The forward wn+1,n and reverse wn,n+1 transition rates between neighboring bins n and n+1 were determined as the trajectory time series of the quantities and bin residence times. To enable a more detailed analysis, NATA coordinates were saved every 0.01 ps. The bin sizes were selected so that only transitions to neighboring bins were observed. Finally, the diffusion coefficients were obtained from the transition rates following Eqns. A3 and A5 from57 with bin width d:

Dn+1/2

 p (n)  = d 2 wn+1,n  eq   peq (n +1) 

1/2

Preferential coordination was described using Kirkwood-Buff type integrals Gab(R)57,58. First, the excess coordination of site a by site b is defined as R

N (R) = ρbGab (R) = ρb ∫ 4π r 2 ( gab (r) −1) dr ex ab

0



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

This quantity measures the local excess of neighbors of b type in a sphere of radius R around an a site, relative to the average expected from the number density of b, ρb = Nb/V. Here gab(r) is the radial distribution function (RDF) of b around a, Nb is the number of b sites and V the system volume. The preferential coordination of site a by b relative to c is then defined as ex Γ abc (R) = ρb ( Gab (R) − Gc (R)) = N ab (R) −

ρb ex N (R) ρc ac

i.e. it compares the excess coordination of a by species b and c, corrected for their overall densities. If we choose a as a site on the solute (peptide), b as a water oxygen and c as an atom on a co-solvent molecule, Γ tells us which species – water or co-solvent has higher excess local density. Γ is related to the free energy ∆G of transfer of co-solvent from bulk solution to the vicinity of the solute through ∆G = -RT ln Γ , with R – the gas constant and T – the absolute temperature59,60.

Acknowledgements. GSJ would like to thank Attila Szabo for helpful discussion. This project was partly supported by the University of Kansas General Research Fund and supported in part by DOE grant DEFG0208ER46528 (KK). AS was partially supported by NCN grant 2012/05/B/ST8/03010.


Page 18 of 35

Page 19 of 35

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


References. 1. Nozaki, Y.; Tanford, C. The Solubility of Amino Acids and Related Compounds in Aqueous Urea Solutions. J. Biol. Chem. 1963, 238, 4074-4081. 2. Tanford, C. Protein denaturation. C. Theoretical models for the mechanism of denaturation. Adv. Protein Chem. 1970, 24, 1-95. 3. Nozaki, Y.; Tanford, C. The Solubility of Amino Acids, Diglycine, and Triglycine in Aqueous Guanidine Hydrochloride Solutions. J. Biol. Chem. 1970, 245, 1648-1652. 4. Greene, R. F.; Pace, C. N. Urea and guanidine hydrochloride denaturation of ribonuclease, lysozyme, alpha-chymotrypsin, and beta-lactoglobulin. J. Biol. Chem. 1974, 249, 5388-5393. 5. Schellman, J. A. Fifty years of solvent denaturation. Biophys. Chem. 2002, 96, 91-101. 6. Yancey, P. H.; Clark, M. E.; Hand, S. C.; Bowlus, R. D. ; Somero, G. N. Living with water stress: evolution of osmolyte systems. Science 1982, 217, 1214-1222. 7. Privalov, P. L. Cold denaturation of proteins. Crit. Rev. Biochem. Mol. Biol. 1990, 25, 281305. 8. Heremans, K. High pressure effects on proteins and other biomolecules. Annu. Rev. Biophys. Bioeng., 1982, 11, 1-21. 9. Timasheff , S. N. The control of protein stability and association by weak interactions with water: how do solvents affect these processes? Annu. Rev. Biophys. Biomol. Struct. 1993, 22, 6797. 10. Nosworthy, N. J; Ginsburg, A. Thermal unfolding of dodecameric glutamine synthetase: inhibition of aggregation by urea. Protein Science 1997, 6, 2617-2623.

11. Klimov, D. K.; Straub, J. E.; Thirumalai, D. Aqueous urea solution destabilizes Abeta(1622) oligomers. Proc. Natl. Acad. Sci, U.S.A. 2004, 101, 14760-14765. 12. Canchi, D. R.; Garcia, A. E. Cosolvent effects on protein stability. Annu. Rev. Phys. Chem. 2013, 64, 273–293. 13. Wyman, J. J. Linked Functions and Reciprocal Effects in Hemoglobin: A Second Look Adv. Protein Chem. 1964, 19, 223-286.



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

14. Tanford C. Protein denaturation. Adv. Protein Chem. 1968, 23, 121-282. 15. Wallqvist, A.; Covell, D. G. Hydrophobic Interactions in Aqueous Urea Solutions with Implications for the Mechanism of Protein Denaturation. J. Am. Chem. Soc. 1998, 120, 427-428. 16. Shimizu, S.; Chan, H. S. Origins of Protein Denatured State Compactness and Hydrophobic Clustering in Aqueous Urea: Inferences from Nonpolar Potentials of Mean Force. Proteins Struct. Funct. Genet. 2002, 49, 560-566. 17. Oostenbrink, C.; van Gunsteren, W. F. Methane clustering in explicit water: effect of urea on hydrophobic interactions. Phys. Chem. Chem. Phys. 2005, 7, 53-58. 18. Lee, M. E.; van der Vegt N. F. A. Coherent Microscopic Picture for Urea-Induced Denaturation of Proteins. J. Am. Chem. Soc. 2006, 128, 4948-4949. 19. O’Brien, E. P.; Dima, R. I.; Brooks, B.; Thirumalai, D. Interactions between hydrophobic and ionic solutes in aqueous guanidinium chloride and urea solutions: lessons for protein denaturation mechanism. J. Am. Chem. Soc. 2007, 129, 7346-7353. 20. Zangi, R.; Zhou, R. H.; Berne, B. J. Urea's action on hydrophobic interactions. J. Am. Chem. Soc. 2009, 131, 1535-1541. 21. Yang, L. J.; Gao, Y. Q. Effects of cosolvents on the hydration of carbon nanotubes. J. Am. Chem. Soc. 2010, 132, 842-848. 22. Das, P.; Zhou, R. H. Urea-induced drying of carbon nanotubes suggests existence of a dry globule-like transient state during chemical denaturation of proteins. J. Phys. Chem. B 2010, 114, 5427-5430. 23. Frank, H. S.; Franks, F. Structural Approach to the Solvent Power of Water for. Hydrocarbons; Urea as a Structure Breaker. J. Chem. Phys. 1968, 48, 4746-4757. 24. Dill, K. A. Dominant forces in protein folding. Biochemistry 1990, 29, 7133-7155. 25. Soper, A. K.; Castner, E. W.; Luzar , A. Impact of urea on water structure: a clue to its properties as a denaturant? Biophys. Chem. 2003, 105, 649-666. 26. Batchelor, J. D.; Olteanu, A.; Tripathy , A.; Pielak, G. J. Impact of protein denaturants and stabilizers on water structure. J. Am. Chem. Soc. 2004, 126, 1958-1961.


Page 20 of 35

Page 21 of 35

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


27. Rezus , Y. L. A.; Bakker, H. J. Effect of urea on the structural dynamics of water. Proc. Natl. Acad. Sci.U.S.A. 2006, 103, 18417-18420. 28. Kokubo, H.; Pettitt, B. M. Preferential solvation in urea solutions at different concentrations: properties from simulation studies. J. Phys. Chem. B 2007, 111, 5233-5242. 29. Stumpe, M. C.; Grubmuller , H. Aqueous urea solutions: structure, energetics, and urea aggregation. J. Phys. Chem. B 2007, 111, 6220-6228. 30. Hua, L.; Zhou, R. H.; Thirumalai, D.; Berne, B. J. Urea denaturation by stronger dispersion interactions with proteins than water implies a 2-stage unfolding. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 16928-16933. 31. Horinek, D.; Netz, R. R. Can simulations quantitatively predict peptide transfer free energies to urea solutions? Thermodynamic concepts and force field limitations. J. Phys. Chem. A 2011, 115, 6125–6136. 32. Ignatova, Z.; Gierasch, L. M. Inhibition of protein aggregation in vitro and in vivo by a natural osmoprotectant. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 13357-13361. 33. Cardenas, A. E.; Shrestha, R.; Webb, L. J.; Elber, R. Membrane permeation of a peptide: it is better to be positive. J. Phys. Chem. B 2015, 119, 6412-6420. 34. Cardenas , A. E.; Jas, G. S.; DeLeon, K. Y.; Hegefeld, W. A.; Kuczera, K.; Elber, R. Unassisted transport of N-acetyl-L-tryptophanamide through membrane: experiment and simulation of kinetics. J. Phys. Chem. B 2012, 116, 2739-2750. 35. Kuczera, K.; Unruh, J.; Johnson, C. K.; Jas G.S. Reorientations of aromatic amino acids and their side chain models: anisotropy measurements and molecular dynamics simulations. J. Phys. Chem. A 2010, 114, 133-142. 36. Kawahara, K.; Tanford, C. Viscosity and density of aqueous solutions of urea and guanidine hydrochloride. J. Biol. Chem. 1966, 241, 3228-3232.

37. Samuel, D.; Kumar, T. K. S.; Jayaraman, G.; Yang, P.W.; Yu, C. Proline is a protein solubilizing solute. Biochem. Mol. Biol. Intl. 1997, 41, 235-242. 38. Yu, H.T.; Colucci, W.J.; McLaughlin, M.L.; Barkley, M.D. Fluorescence quenching in indoles by excited-state proton transfer. J. Am. Chem. Soc. 1992, 114, 8449-8454.



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

39. Chen, Y.; Liu, B.; Yu, H.T.; Barkley, M.D. The Peptide Bond Quenches Indole Fluorescence. J. Am. Chem. Soc. 1996, 118, 9271-9278. 40. Guevara-Carrion, G.; Vrabec, J.; Hasse, H. Prediction of self-diffusion coefficient and shear viscosity of water and its binary mixtures with methanol and ethanol by molecular simulation. J. Chem. Phys. 2011, 134, 074508. 41. Luzar, A.; Chandler D. Hydrogen-bond kinetics in liquid water. Nature 1996, 379, 55-57. 42. Bicout D. J.; Szabo, A. Electron transfer reaction dynamics in non-Debye solvents. J. Chem. Phys. 1998, 109, 2325-2338. 43. Wallqvist, A.; Covell, D. G.; Thirumalai, D. Hydrophobic Interactions in Aqueous Urea Solutions with Implications for the Mechanism of Protein Denaturation. J. Am. Chem. Soc. 1998, 120, 427-428. 44. Tsai, J.; Gerstein, M.; Levitt, M. Keeping the Shape but Changing the Charges: A Simulation Study of Urea and its Isosteric Analogues J. Chem. Phys. 1996, 104, 9417-9430. 45. Rezus, Y. L. A.; Bakker, H. J. Effect of urea on the structural dynamics of water. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 18417-18420. 46. Mountain, R. D.; Thirumalai, D. Alterations in Water Structure Induced by Guanidinium and Sodium Ions. J. Phys. Chem. B 2004, 108, 19711-19716. 47. Mason, P. E.; Dempsey, C. E.; Neilson, G. W.; Kline, S. R.; Brady, J. W. Preferential interactions of guanidinum ions with aromatic groups over aliphatic groups. J. Am. Chem. Soc. 2009, 131, 16689-16696. 48. Bolen, D. W. & Baskakov, I. V. The osmophobic effect: natural selection of a thermodynamic force in protein folding. J. Mol. Biol. 2000, 310, 955-963. 49. Auton, M. & Bolen, D. W. Predicting the energetics of osmolyte-induced protein folding/unfolding. Proc. Natl. Acad. Sci. USA 2005, 102, 15065-15068. 50. Harms, G. S.; Pauls, S. W.; Hedstrom, J. F.; Johnson, C. K. Tyrosyl Fluorescence Decays and Rotational Dynamics in Tyrosine Monomers and in Dipeptides. J. Fluoresc. 1997, 7, 273-282.

51. Shen, X.; Knutson, J. R. Subpicosecond Fluorescence Spectra of Tryptophan in Water J. Phys. Chem. B 2001, 105, 6260-6265.


Page 22 of 35

Page 23 of 35

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


52. Brooks, B. R.; Brooks, C. L.; Mackerell, A. D.; Nilsson, L.; Petrella, R. J.; Roux, B.; Won, Y.; Archontis, G.; Bartels, C.; Boresch, S. et al. CHARMM: The Biomolecular Simulation Program. J. Comput. Chem. 2009, 30, 1545-1614. 53. Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable Molecular Simulation J. Chem. Theory Comput. 2008, 4, 435-447. 54. Kaminski, G. A.; Friesner, R. A.; Tirado-Rives, J.; Jorgensen, W. L. Evaluation and Reparametrization of the OPLS-AA Force Field for Proteins via Comparison with Accurate Quantum Chemical Calculations on Peptides. J. Phys. Chem. B 2001, 105, 6474-6487.

55. Carpenter, G. A.; Grossberg, S.; Rosen, D.B. ART 2-A: An adaptive resonance algorithm for rapid category learning and recognition. Neural Networks 1991, 4, 493-504. 56. Karpen, M. E.; Tobias, D. T.; Brooks III, C. L. Statistical clustering techniques for the analysis of long molecular dynamics trajectories: analysis of 2.2-ns trajectories of YPGDV. Biochemistry 1993, 32, 412-420. 57. Luzar A. Resolving the hydrogen bond dynamics conundrum. J. Chem. Phys. 2000, 113, 10663-10675. 58. van der Spoel D.; van Maaren, P. J.; Larson, P.; Timneanu, N. Thermodynamics of hydrogen bonding in hydrophilic and hydrophobic media. J. Phys. Chem. B 2006, 110, 4393-4398. 59. Pierce, V.; Kang, M.; Aburi, M.; Weerasinghe, S.; Smith, P. E. Recent applications of Kirkwood-Buff theory to biological systems. Cell Biochem Biophys. 2008, 50, 1-22.

60. Smiatek, J. Osmolyte effects: impact on the aqueous solution around charged and neutral spheres. J. Phys. Chem. B 2014, 118, 771-782.



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

Figure 1.


Page 24 of 35

Page 25 of 35

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


Figure 2.



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

Figure 3.


Page 26 of 35

Page 27 of 35

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


Figure 4.



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

Figure 5.


Page 28 of 35

Page 29 of 35

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


Figure 6.



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

Figure 7.


Page 30 of 35

Page 31 of 35

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


Figure 8.



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

Figure 9.


Page 32 of 35

Page 33 of 35

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


Figure 10.



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

Figure 11.


Page 34 of 35

Page 35 of 35

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


TOC.


Reorientation Motion and Preferential Interactions ... - ACS Publications

Recommend Documents