Probing Selection Mechanism of the Most Favorable Conformation of

Subscriber access provided by The Chinese University of Hong Kong

Article

Probing Selection Mechanism of the Most Favorable Conformation of a Dipeptide in Chaotropic and Kosmotropic Solution Gouri S. Jas, Charles Russell Middaugh, and Krzysztof Kuczera J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.6b04528 • Publication Date (Web): 21 Jun 2016 Downloaded from http://pubs.acs.org on June 24, 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 39

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

Probing Selection Mechanism of the Most Favorable Conformation of a Dipeptide in Chaotropic and Kosmotropic Solution. Gouri S. Jas1*, C. Russell Middaugh1, Krzysztof Kuczera2,3 1

Department of Pharmaceutical Chemistry, The University of Kansas, Lawrence, KS 66047 2 3

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

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

*Gouri S. Jas Department of Pharmaceutical Chemistry The University of Kansas, Lawrence, KS 66047 785-218-5186 [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

Page 2 of 39

Abstract. Chaotropes like urea and guanidinium chloride (GdmCl) tend to destabilize, and kosmotropes like proline tend to stabilize folded structures of peptides and proteins. Here, we combine fluorescence anisotropy decay measurements and molecular dynamics simulations to gain a microscopic understanding of the molecular mechanism for shifting conformational preferences in aqueous, GdmCl, urea, and proline solutions of a simple model dipeptide, Nacetyl-tryptophan-amide (NATA). Measured anisotropy decay of NATA as a function of temperature, pH and co-solvent concentrations showed reorientations moderately slower in GdmCl and urea and substantially slower in proline compared to aqueous environment. A small change in pH significantly slows orientation time in water and GdmCl and less markedly in urea. Computationally, we use molecular dynamics with dihedral restraints to separately analyze the motions and interactions of the representative NATA conformers in the four different solvent environments. This novel analysis provides a dissection of the observed overall diffusion rates into contributions from individual dipeptide conformations. The variation of rotational diffusion rates with conformation are quite large. Population-weighted averaging or using properties of the major cluster reproduces the dynamical features of the full unrestrained dynamics. Additionally, we correlate the observable diffusion rates with microscopic features of conformer size, shape and solvation. This analysis uncovered underlying differences in detailed atomistic behavior of the three co-solvents – urea, GdmCl and proline. For both urea and the pure water system we find good agreement with hydrodynamic theory, with diffusion rates primarily correlated with conformer size and shape. In contrast, for GdmCl and proline solutions, the variation in conformer diffusion rates was mostly determined by specific interactions with the co-solvents. We also find preferences for different molecular shapes by the three co-solvents, with increased preferential solvation of smaller and more spherical conformers by urea and larger and more elongated conformers by GdmCl and proline. Additionally, our results provide a basis for a simple approximate model of the effects of pH lowering on dipeptide conformational equilibria. The translational diffusion rates of NATA are less sensitive to conformations, but variation with solvation strength is similar to rotational diffusion. Our results, combining experiment and simulation, show that we can identify the individual peptide conformers with definite microscopic properties of shape, size and solvation, that are responsible for producing physical observables such translational and orientational diffusion in the complex solvent environments of denaturants

and


osmolytes.

Page 3 of 39

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. In solution, the folding equilibrium of peptides and proteins can be shifted due to the inclusion of small organic molecules such as urea, guanidinium chloride (GdmCl), and proline1,2. These organic co-solvents, when moving the equilibrium toward the native state are called osmolytes or kosmotropes. Co-solvents that shift the chemical equilibrium to the more unfolded state are known as denaturants or chaotropes. At ambient conditions peptides and proteins exist in conformational equilibrium with an ensemble of preferred folded state with unfolded states. By altering the thermodynamic state of the system with temperature, pH or by the addition of cosolvents this thermodynamic equilibrium can be perturbed, resulting in a shift of conformer populations3. The denaturation process, a shift of the conformational equilibrium to the nonnative form, occurs naturally in the tissue of marine elasmobranchs fish under stressful conditions. The fish counteract this effect by producing protective osmolytes and stabilizing the native state4. Protein denaturation by urea has been studies extensively in order to understand the molecular mechanism that drive this process5-8. Urea has been shown to preferentially congregate around proteins and interact favorably with the backbone as well as the side chains9. The largely favorable preferential interaction of urea with the unfolded protein ensemble is the thermodynamic driving force behind unfolding. Much less is known about the molecular origin of protein denaturation by guanidinium chloride (GdmCl). One of the interesting aspects of GdmCl-denatured states is their structural shape. It has been suggested that urea tend to expand the denatured state9. Not much is known about the structural shape of the denatured state in GdmCl. Naturally occurring osmolytes such proline is known to stabilize the native state and thus shift the equilibrium toward the folded form. The detailed molecular mechanism of the stabilization process is by the osmolytes on protein stability is not well characterized. It has been shown that osmolyte, such as Trimethylamine N-oxide (TMAO) can fold thermodynamically unfolded proteins to significantly functional native-like structures10,11. Based on transfer free energy measurements, it has been suggested that both classes of co-solvents influence the protein through the peptide backbone, whereas denaturants favorably and osmolytes unfavorably interact with the backbone12-14. A clear convincing picture of this proposed model is yet to emerge. Urea denaturation has been studied extensively with molecular dynamics, for both simple model systems and proteins. In order to separate charge and polarity from purely hydrophobic effects,



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 39

model systems were studied. Conflicting results emerged related to weakening of the hydrophobic effect due to addition of urea to small hydrophobic solutes15-18. The convincing effect of urea has been clearly observed in simulations of model hydrophobic systems at a longer time scale. It has shown that the attraction between hydrophobic plates is reduced in urea because of the stabilization of the liquid phase between the plates against dewetting19. A purely hydrophobic polymer is shown to unfold in urea as result of enthalpically favorable dispersion interactions of polymer with urea20. These studies show that urea can modulate the hydrophobic effect by making energetically favorable contacts with the purely hydrophobic groups. Compared to urea, studies of denaturation of model system and proteins in GdmCl are not very common. The rates of nanosecond and sub-nanosecond of molecular motions and interactions with the immediate environment can be directly studied by measuring molecular reorientation in solution 21-23

. The interpretation of the orientation motion in a complex solvent environment of a simple

model peptide such as N-acetyl-tryptophan-amide (NATA) can be very useful starting point for describing the nature of interaction with the environment and the influence of co-solvent on the structural shape of larger systems such as helices, hairpins, and proteins. One of the important components the present 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. In terms of continuum hydrodynamics, the dependence of diffusion rates on molecular shape and size and solvent type is generally well understood. For a spherical solute under stick conditions the translational diffusion coefficient Dt is

Dt =

kT 6πη R

and the rotational diffusion coefficient Drot and the corresponding rotational correlation time τrot are

Drot =

kT 6ηV

; τ rot =

1 ηV = 6Dr kT

with k the Boltzmann constant, T the absolute temperature in Kelvin, η the solvent viscosity, R the solute radius and V its volume

24.

Analytical results have been obtained for ellipsoids

24-26

,

but currently effective numerical approaches are available for arbitrarily shaped molecules 27. As


Page 5 of 39

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


indicated by the above equations, in the hydrodynamic limit, the diffusion rates are inversely proportional to viscosity and also inversely proportional to solute size (Dt) or volume (Dror). For objects with same volume, deviation from spherical shape leads to slower diffusion, though for ellipsoids with long/short axis ratios of 2 the increases in Dr and Dt are only about 4% compared to a sphere 24. In our previous study of NATA in the presence of co-solvents, we have found interesting deviations from the simple continuum model presented above23. Namely, the experimentally measured rotational correlation times of NATA in 5.5 M solutions of urea, GdmCl and proline were 2.4, 3.3 and 10 times higher compared to water, while the corresponding solution viscosity ratios were 1.3, 1.7 and 7 at room temperature. This led us to suggest that besides viscosity effects, specific solute-solvent interactions play an important role in determining NATA reorientation rates in presence of co-solvents. These interactions were analyzed using all-atom molecular dynamics simulations. The computational study found very similar diffusion rates to the experimental data. Additionally, simulations showed that urea interacts strongly but not specifically with both backbone and sidechain parts of NATA, while guanidinium and proline interact preferentially and in a very specific manner with the sidechain. Thus, using solvation patterns we were able to explain the unusually strong slowdown of reorientation rates of NATA in urea, GdmCl and proline solutions. Additionally, the MD simulations showed that NATA is a flexible molecule, sampling multiple conformers with populations varying in a subtle way with the co-solvent type23. In this work, we combine fluorescence anisotropy decay measurements and all atom molecular dynamics simulations to gain a deeper understanding of the molecular mechanisms that drive the conformational preferences of the model NATA dipeptide in aqueous, GdmCl, urea, and proline solution. Experimentally, we present new measurements of NATA fluorescence lifetime, rotational anisotropy at pH 4.8 and four different temperatures ranging 280-310K. In this study we perturb the thermodynamic equilibrium by slightly lowering the pH and interestingly, this small change in pH significantly slows reorientation times in aqueous and GdmCl solutions, but not as significantly in urea. Computationally, we use molecular dynamics with dihedral restraints to separately analyze the motions and interactions of the representative NATA conformers identified previously23. For each co-solvent system that gives us a sample 8-10 dipeptide



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 39

structures of known shape and size, for which we calculate individual diffusion rates and solvation patterns. This novel analysis provides a dissection of the observed overall diffusion rates into contributions from individual dipeptide conformations. Interestingly, while variation of Dt with conformation is quite small, the variation of Drot and τrot are quite large. Additionally, our sets of individual conformers provide us with data that allow for correlations of the observed features - Dt and τrot – with microscopic determinants – size, shape and solvation. This analysis throws new light on the difference in detailed atomistic behavior of the three co-solvents – urea, GdmCl and proline. Additionally, our results provide a basis for a simple model of the effects of pH lowering on dipeptide conformational equilibria.

Methods. Experimental methods Materials. NATA, N-acetyl-tryptophan-amide, was purchased from Sigma-Aldrich with the purity > 99%. Anisotropy measurements are conducted in 20 mM acetate buffer at pH 4.8 with a 30-40µM sample concentration. Denaturant and osmolyte solutions were made in 20mM acetate buffer. Fluorescence anisotropy decay measurements. Experimental technique is explained in detail elsewhere. Briefly, 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 signal of was collected at the fluorescence maxima, 350nm with an 8-nm bandpass (model 9030 monochromator, Sciencetech Inc, Concord, ON, Canada). As described in detail elsewhere23,28,

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). Fluorescence decays were measured for samples in aqueous, urea, GdmCl, and proline from 0 to 5.5M at 280 K, 290 K, and 300, 310, 320 K. 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


Page 7 of 39

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


by globally fitting the fluorescence decays with polarizations parallel and perpendicular to the excitation polarization as described elsewhere29. The initial anisotropy r(0) at t=0 is kept fixed at 0.25 in the fitting procedure as described elsewhere30. In each case the anisotropy decay was fit with a two-exponential decay. Simulation methods. The simulated peptide system, N-acetyl-tryptophan-amide, Ac-Trp-NH2 (NATA) was studied in the indicated co-solvents. The starting unrestrained 100 ns trajectories for NATA in TIP4P water, 5M urea solution and 5M proline solution were as described previously23, with Namidated and C-acetylated termini. Compared to the previous work, a new unrestrained trajectory of NATA with TIP4P water in 5 M GdmCl was generated, using the Kirkwood-Buff force field for guanidinium 31. A pre-equilibrated box of 5 M GdmCl with TIP4P was generated with GROMACS and used to overlay NATA, producing a system including NATA, 92 guanidinium chloride ion pairs and 869 TIP4P waters. A brief energy minimization and short simulation with restrained NATA was followed by a 1 ns equilibration at 300 K under NPT conditions at 1 atm pressure. Finally, a 100 ns MD trajectory for NATa in GdmCL/KBFF was generated at 300 K under NVT conditions. The equilibrated box size was 33.3 Å.

For the four 100 ns unrestrained MD trajectories at 300 K, clustering was performed in dihedral space, using the time series of φ, ψ, χ1 and χ2 dihedrals sampled every 1 ps. The ART-2’ algorithm implemented in the CHARMM program was used

32,33

. With a 90o radius, eight

clusters overall were identified in water, urea and proline, as described before23 and ten in the new GdmCl trajectory. For each of the 34 clusters, a central structure was extracted from the corresponding trajectory, including solvent and co-solvent, and an MD trajectory with restrained φ, ψ, χ1 and χ2 dihedrals, with reference set to the cluster center, was used to start a 50 ns NVT MD trajectory. A force constant of 100 kJ mol-1 rad-2 was used for the restraints, which yielded trajectories with root-mean-square dihedral fluctuations of about 8o.

NATA structures

representing the clusters are presented in Figure 1, while the cluster populations are described in Figure 2. A more complete description of the clusters is given in Supplementary Information Table S5.



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 39

All simulations used 14 Å van der Waals cutoffs and 10 Å direct cutoffs with PME for electrostatics. The GROMACS version 4.5.6 program34 was employed, using a 2 fs time step and constraints on all bond lengths, with the OPLS/AA force field 35. For each of the 34 restrained trajectories we calculated the NATA rotational correlation time, translational diffusion coefficient and solvation patterns. The choice of force field is motivated by two factors. First, in our previous simulations of short model peptides use of OPLS/AA parameters led to very good agreement on helix content and folding kinetics with experimental data

36-38

. Second, to obtain

diffusion coefficients comparable to experimental data, we need to employ a water model with reasonable viscosity and self-diffusion, such as TIP4P 39.

Molecular reorientations in MD trajectories were analyzed by generating time series of transition dipole axis for the NATA side-chain 1Lb transition28 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 peptide centerof-mass to a straight line in the 100-1100 ps intervals. Solvation patterns were analyzed through radial distribution functions, excess coordination and preferential coordination. The excess coordination of site a by site b is defined using KirkwoodBuff type integrals Gab(R)40,41 R

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

0

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 ac (R) ρc


Page 9 of 39

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


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 relative 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 temperature 40,41. To characterize the solute size and shape in the individual conformers, the radii of gyration Rg, shape parameters δ*9 and the ratios of long to short molecular axes x/z were calculated

with

CHARMM42 for the 34 NATA starting structures.

Results. Fluorescence Anisotropy Decay. Fluorescence anisotropy decay measurements are designed to probe directly the solute reorientational dynamics and its interaction with the immediate environment. NATA, a simple model dipeptide system, with two peptide bonds, is subjected to this measurement to explore the change in the nature of dipeptide dynamics and interactions with respect to its current surroundings. In this work we have chosen two chaotropes, urea and GdmCl and one kosmotrope, proline, with a range of concentrations upto 5.5M and in a temperature range of 280-320K. The thermodynamic equilibrium of the dipeptide solution in chaotropes was perturbed with another variable, a change in pH (from pH 7.2 to pH 4.8) to examine the effect on the orientation dynamics and explore the resulting change in interaction with the environment. We are especially interested in separating out effects of bulk (hydrodynamic) properties from specific solvation. As noted in the introduction, bulk effects depend in a simple way on solvent viscosity. At 298K the measured solvent viscosities are ~1.3 cp for 5M urea-water, ~1.5 cp in 5M GdmCl-water, ~7 cp in 5M proline-water, and ~1cp in the pure aqueous system23. A summary plot describing the fluorescence lifetime and global orientation time of NATA in aqueous, urea, and GdmCl solution is presented in Figure 3. Representative fluorescence lifetime measurements in aqueous solution, and in presence of chaotropes and kosmotropes, and



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 39

fluorescence intensities in parallel and perpendicular polarization at 300K in the four different solvent conditions are presented in Figures 4(A-D) and 4(E-H), respectively. Anisotropy decay plots of NATA in four different solvent conditions are presented in Figure 5. Figures 5(A-D) show the anisotropy decay with a 2-exponential fit to the decay curve, Figures 5(E-H) show the summary of orientation times in two different pHs. NATA fluorescence is observed to be substantially quenched in the presence of proline co-solvent, compared to other three solvent conditions, possibly due to proton and/or electron transfer. This quenching behavior does not influence the anisotropy as it is determined from the intensity ratio of measured fluorescence spectra. Computed C2(t) for NATA in aqueous, urea, GdmCl, and proline are presented in Figure 6(A-D). Orientation time obtained from computed C2(t) are found to be in a very good agreement with the measured data in all four solvent conditions. Measured orientation times, in two different pHs, with respect to the orientation time of specific conformers are shown in figure 6 (E-F).

A summary of the structural shapes of conformers (shown in Figures 1A-D) and

corresponding free energies are shown in Figures 2A-D. Measured rotational orientation times (τrot) determined from fluorescence anisotropy are found to be slower by a factor of 2.4 in 5M urea, 3.8 in 5M GdmCl, and 10.5 in 5M proline compared to water at 300 K. Measured cosolvent viscosities of 5M urea, GdmCl and proline are higher by factors of about 1.3, 1.5 and 7, respectively, compared aqueous solution at 300K. A pure viscosity effect is unable to account for this significantly slower orientation time in these co-solvents, even with higher viscosities of these co-solvents compared to water. One of the striking observations came out of our decay analysis is that the global orientation time of NATA in 5M proline is slower by a factor of 10.5. On the other hand, the co-solvent viscosities are slightly higher than water in urea and GdmCl (1.3 and 1.5) but the rotational orientation time is observed to be slower by a factor of 2.4 in urea and 3.8 in GdmCl, respectively. At 300K and pH 4.8, NATA in aqueous and GdmCl produced a significantly slower rotational orientation time compared to pH 7.2. It aqueous solution it observed to be slower by a factor of 1.3 and in GdmCl, under identical GdmCl concentration, it is slower by a factor 1.2. In urea, on the other hand, NATA orientation time was very similar to pH 7.2 (56 ps in pH 7.2, 60 ps in pH 4.8). This may suggest that there may not be a significant change in the dominant population of NATA conformers in urea under these two pHs.


Page 11 of 39

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


Restrained molecular dynamics simulations The representative conformers for NATA were obtained by clustering unrestrained trajectories of the systems, as described in Methods and elsewhere23. Next, molecular dynamics simulations with dihedral restraints were performed for the individual conformers populated by NATA in the four studied environments – eight in water, eight in 5 M urea, ten in 5 M GdmCl and eight in 5 M proline. The simulation results, describing the dipeptide dynamics and solvation for the contributing structures, are presented below. Rotational diffusion. The rotational correlation times τrot of the individual conformers were in the 28-42 ps range in water, 40-57 ps in urea, 68-115 in GdmCl and 230-443 ps in proline. The τrot values thus exhibited a wide range of variation, from 40-50% in water and urea to 70-90% in GdmCl and proline (Figure 7). Table 1 compares the τrot values of the conformers to the computational and experimental results for the unrestrained peptide. The population-weighted averages over conformer results are within statistical error of the unrestrained trajectory calculations. This indicates that our approach of dissecting peptide dynamics into conformer contributions is consistent, as synthesis of the conformer data essentially reproduces the overall peptide dynamics. The success of the method may be rationalized by time-scale separation – while reorientations occur on 10-100 ps times, conformer transitions are in the nanosecond regime

22

. As shown in Table 1, using the data from restrained simulations of the most highly

populated conformer only, also leads to reasonable agreement with the unrestrained simulation and experimental data. τrot Free MD Weighted average Major conformer Experiment

Water 30 33 28 30

Urea 44 46 40 60

GdmCl 86 94 105 98

Proline 340 310 230 304

Table 1. Rotational diffusion of NATA conformers – τrot in ps. Free MD - τrot for NATA at 300 K from 100 ns unconstrained simulations23. Weighted average – population-weighted average, with weights from cluster populations in the unrestrained MD and conformer correlation times from simulations with restrained dihedrals (Fig. 7). Major conformer – τrot value from restrained MD for conformer representing cluster with highest population in the corresponding unrestrained trajectory. The statistical errors of the τrot values are 4-6% for water, urea and GdmCl and 10-12% for proline.



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 39

Translational diffusion. The translational diffusion coefficients of the individual conformers were in the 0.69-0.79×10-9 m2s-1 range in water, 0.42-0.57×10-9 m2s-1 in urea, 0.28-0.36 ×10-9 m2s-1 in GdmCl and 0.10-0.14×10-9 m2s-1 in proline (Figure 8). Thus, the translational diffusion rate exhibited a relatively small range of variation (about 20%) in water and moderate range variation (30-40%) in the presence of the three co-solvents. Table 2 compares the Dt values of the conformers with those of the unrestrained peptide. The population-weighted average of the conformer Dt values agrees with the unconstrained diffusion coefficient within statistical errors. This indicates that our approach of dissecting overall dynamics into components from individual conformers leads to a consistent picture of peptide translational motions. Using the value of the most highly populated conformer also yields results similar to those from the unconstrained trajectory. Thus, a consistent picture of both NATA translational and rotational diffusion may be obtained by dissecting overall peptide dynamics into contributions from individual conformers. Using the samples of structures and motions presented by the conformer trajectories, we can further analyze their behavior by correlating diffusion rates with shape, size and solvation.

Dt Free MD Weighted average Major conformer

Water 0.80 0.76 0.77

Urea 0.61 0.51 0.53

GdmCl 0.31 0.32 0.31

Proline 0.11 0.12 0.13

Table 2. Translational diffusion of NATA conformers (units: 10-9 cm2/s). Free MD - Dt for NATA at 300 K from 100 ns unconstrained simulations 23. Weighted average – population-weighted average, with weights from cluster populations in the unrestrained MD and conformer diffusion coefficients from simulations with restrained dihedrals (Fig. 8). Major conformer – Dt value from restrained MD for conformer representing cluster with highest population in the corresponding unrestrained trajectory. The statistical errors of the Dt values are 10-20%.

Peptide shape and size The size dependence of the conformer diffusion rates is presented in Figures 7 and 8. First we must note that the change in average peptide size, as measured by the radius of gyration Rg, varies only slightly between NATA conformers, typically in the 3.05-3.45 Å range. The rotational correlation time tends to increase with conformer size in water and urea, while dependence on Rg is weak for GdmCl and proline. In contrast, the translational diffusion rate appears mostly insensitive to conformer radius of gyration in all four solvents.


Page 13 of 39

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


Two measures have been used to describe peptide shape: δ*, which varies between 0 for spherical to 1 for elongated shapes (figures 9 and 10) and x/z, the ratio of longest to shortest principal axes of the molecule (figures 11 and 12). The data show a definite trend for increase of τrot for the more elongated structures (figures 9 and 11). The trend is strongest in water and urea, somewhat weaker in GdmCl and weakest of all in proline. The correlation of Dt with conformer elongation is mostly quite weak, except in the case of proline, which exhibits a slight tendency for slower translation diffusion with structure elongation. Peptide solvation Slightly different measures of interaction with solvent were employed in the four studied systems. In the water simulation, in which there are no co-solvents, we used the excess coordination of the indole sidechain by water. In the urea system we employed preferential coordination of the indole by urea nitrogens and for GdmCl the preferential coordination of indole by guanidinium nitrogens. Finally, for proline, we used excess coordination of indole by proline Cγ atoms, which was found to be a very strong effect in the unconstrained simulations23. The rotational correlation times exhibited moderate sensitivity to these descriptors of solvation. There is a clear tendency for increase in τrot with solvation in water, GdmCl and proline, while a decrease in τrot with solvation strength was detected for urea (Fig 13). For translational diffusion the effect of solvation was very weak in water, a decrease in Dt with solvation strength was found in GdmCl and proline, and an increase in urea (Fig. 14). Solvation vs. shape and size The correlations between the solvation descriptors and peptide shape and size are shown in Fig. 15. In the water system, increased solvation of indole by water strongly correlates with both larger conformer size and greater elongation. Given the strong negative correlation of NATA diffusion rates in water with size and shape, this explains the tendency of increasing τrot and decreasing Dt with water coordination.

In GdmCl there is a similar but weaker effect of

increased indole interactions with guanidinium correlating with larger conformer size and elongation. In urea, there is an opposite effect, with a moderate trend for stronger interactions of indole with urea in conformers that are more spherical and smaller in size. For proline essentially no correlation exists between the solvation parameter and shape or size.



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 39

Representative structures and model of pH change A simplified way for describing NATA properties through conformer contributions is by representing the dipeptide by the single conformer with the highest population. Our simulations represent behavior at neutral pH, i.e. correspond to the experimental observations at pH 7.2. Under these conditions the representative conformers are structure #2 in water (Nw2), #1 in urea (Nu1), #6 in GdmCl (Ng6) and #2 in proline (Np2) – see figure 1 for structures and figure 2 for free energy levels. Among conformers sampled in aqueous solution, Nw2 exhibits the lowest excess solvation, smallest size, most spherical shape and fastest correlation time τrot = 28 ps. In the urea systems, Nu1 showed average specific coordination by urea, with smallest size, most spherical shape and slowest orientation time τrot = 40 ps. Conformer Ng6 in GdmCl has the second-highest preferential coordination by guanidinium, second-largest size and second-most elongated shape in its group and τrot = 105 ps, at the high end of the range. The NATA structure Np2 in proline solution exhibits smallest size, most spherical shape and average preferential interaction with proline co-solvent among all of corresponding conformers, as well as τrot = 230 ps, the lowest correlation time. These representative structures have rotational correlation times close to the experimentally determined values (Table I). A simple qualitative model that can explain the rotational correlation time shifts with lowering pH involves a shift in conformational equilibrium involving the structures most highly populated at pH 7.2. In aqueous solution experimental τrot changes from 30 ps at pH 7.2 to 36 ps at pH 4.8. This can be achieved by a population decrease of the Nw2 conformer, which has the lowest τrot = 28 ps, and increase in population of the others. A good candidate for the representative structure at pH 4.8 is thus be Nw1, the second-most populated form, with τrot = 37 ps. In urea, measured NATA reorientation times are 60 ps at pH 7.2 and 56 ps at pH 4.8. A decrease in rotational time upon pH lowering may be brought about by increasing the populations of Nu1 and Nu4, the two conformers with lowest values of τrot ≈ 40 ps, at the expense of the other structures. The representative structure of NATA at pH 4.8 could thus be either Nu1 or Nu4. In GdmCl the measured reorientation times of NATA are 98 ps at pH 7.2 and 115 ps at pH 4.8. The increase of τrot with pH drop may be obtained by either decreasing the populations of most stable structures with rotational times lower than that of Ng6 (such as Ng2 and Ng3) or increasing population of Ng6 and other conformers with higher values of τrot (e.g. Ng4). The representative structure of NATA in GdmCl solution could thus be either Ng6 or Ng4. NMR studies of acetylated


Page 15 of 39

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


aminoacids indicate solvation by one additional water molecule for the protonated (low pH) form compared to the neutral form

43

. This is in qualitative agreement with our modeling results in

water, where the lower pH model Nw1 exhibits about 0.5 more waters of solvation that the higher pH model Nw2 (see Supplementary Information, Table S1).

Discussion The advantage of our dihedral-restrained MD approach is the ability to describe structures, motions and interactions of the individual conformers of NATA. For each environment – water, urea, GdmCl and proline – all corresponding conformers move in a solvent with the same bulk properties. However, each conformer is characterized by a different set of microscopic descriptors of size, shape and solute-solvent interactions. Correlation of motions with the microscopic descriptors allows us to determine the relative roles of the different factors in peptide dynamics. Several interesting conclusions may be drawn from our simulation results, underscoring the different properties of the three co-solvents used. The rotational diffusion rate of NATA is strongly sensitive to conformation – e.g. in the proline system we find τrot values in a 230-443 ps range for the individual structures. However, using either a population-weighted average or the properties of the most populated cluster, we can consistently describe the diffusion rates found previously in unrestrained simulations. The variation of τrot with solute microscopic parameters is markedly different in water and urea vs. GdmCl and proline. In water and urea there is a strong positive correlation between τrot and both conformer size and elongation, i.e. these systems exhibit behavior close to hydrodynamic predictions. In contrast, GdmCl and proline exhibit strong deviations from simple hydrodynamic models, with little or no dependence of diffusion rates on conformer shape and size. A mirror image of this behavior is found in analyzing effects of interactions with solvent. The case of the aqueous solution is simplest, with correlation times increasing with strength of solvation by water, due to improved solvation of structures of larger size and elongation. Urea shows unexpected behavior, a trend for decreasing τrot (i.e. faster rotational diffusion) with stronger indole-urea interactions. This may be explained by the property of the urea solution to better interact with molecules of smaller size and more spherical shape, which could be related to the ability of urea to interact with both backbone and sidechain of NATA. In GdmCl specific



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 39

solvation is the dominant factor influencing rotational diffusion, with a strong positive correlation between preferential solvation and τrot. Similarly, for proline solvation is the major effect, but with weaker correlation. Previously, we found that urea is able to interact favorably but non-specifically with both NATA backbone and sidechain, while guanidinium interacts non-specifically and proline highly specifically with the sidechain23. Our detailed analysis of conformer dynamics has uncovered two novel effects. First, we find that solvation of NATA by water and urea may be well described by continuum hydrodynamic models, while effects of GdmCl and proline cannot. For these two latter co-solvents, specific solute-solvent interactions are the main factor influencing the diffusion rate. This is in accord with the experimentally observed significant slowdown of molecular reorientations of NATA in GdmCl and proline solution, well beyond simple viscosity effects. The second effect is the preference for different molecular shapes by the three cosolvents. Thus, in the case of NATA, we find increased preferential solvation of smaller and more spherical conformers by urea and larger and more elongated conformers by GdmCl and proline. The translational diffusion rates of NATA are less sensitive to conformations. Variation of Dt values with peptide size and shape are small and mostly do not exhibit definite trends. Interestingly, the changes of Dt with solvation strength are qualitatively similar to those of Drot = 1/(6τrot). Namely, the translational diffusion rates show a moderate increase with preferential solvation in urea and moderate decreases with preferential solvation in GdmCl and proline. The conformer rotational relaxation times have relatively broad distributions, 28-42 ps in water, 40-57 ps in urea, 68-115 in GdmCl and 230-443 ps in proline. This suggests a simple approximate model for describing pH effects on τrot values. Namely, we propose that lowering the pH influences the populations of individual conformers, increasing population of conformers with low τrot and decreasing populations of conformers with high τrot, thus producing the observed shifts of NATA rotational correlation times with pH. Thus, we can identify individual conformers responsible for observed spectroscopic signals under specific environmental conditions. The experimental data indicate a strong dependence of reorientation rates on cosolvent concentration. In the simulations we focus on a single concentration of 5 M, as this is in the upper range of experimental data and in the case of chaotropes corresponds to denaturing


Page 17 of 39

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


conditions for most proteins. The properties of co-solvent solutions with lower concentrations in an interesting topic for future studies.

Conclusions We combine fluorescence anisotropy decay measurements and dihedral-restrained molecular dynamics simulations to gain a deeper understanding of the molecular mechanisms that drive the conformational preferences of the model NATA dipeptide in aqueous, GdmCl, urea, and proline solution. Experimentally, we present new measurements of NATA rotational anisotropy at pH 4.8, lower than the previously described results. Interestingly, the small change in pH significantly slows reorientation times in aqueous and GdmCl solutions, but not as significantly in urea. Computationally, we use molecular dynamics with dihedral restraints to separately analyze the motions and interactions of the representative NATA conformers identified previously

23

. For each co-solvent system that gives us a sample 8-10 dipeptide structures of

known shape and size, for which we calculate individual diffusion rates and solvation patterns. This novel analysis provides a dissection of the observed overall diffusion rates into contributions from individual dipeptide conformations. Interestingly, while variation of Dt with conformation is quite small, the variation of Drot and τrot are quite large. Population-weighted averaging or using properties of the major cluster reproduces the dynamical features of the full unrestrained dynamics. Additionally, our sets of individual conformers provide us with data that allow for correlations of the observable features - Dt and τrot – with microscopic determinants, size, shape and solvation. This analysis throws new light on the difference in detailed atomistic behavior of the three co-solvents – urea, GdmCl and proline. First, for both urea and the pure water system we find good agreement with hydrodynamic theory, with diffusion rates primarily correlated with conformer size and shape. In contrast, for GdmCl and proline solutions, the variation in conformer diffusion rates was mostly determined by specific interactions with the co-solvents. Second, there is a preference for different molecular shapes in the three co-solvents. Thus, in the case of NATA, we find increased preferential solvation of smaller and more spherical conformers by urea and larger and more elongated conformers by GdmCl and proline. Additionally, our results provide a basis for a simple approximate model of the effects of pH lowering on dipeptide conformational equilibria.



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 translational diffusion rates of NATA are less sensitive to conformations. Variation of Dt values with peptide size and shape are small and mostly do not exhibit definite trends. However, the changes of translational diffusion rates with solvation strength are qualitatively similar to those of rotational diffusion. Our results, combining experiment and simulation, show that we can identify the individual peptide conformers with definite microscopic properties of shape, size and solvation, that are responsible for producing physical observables such translational and orientational diffusion in the complex solvent environments of denaturants and osmolytes. The insightful microscopic explanations for the unexpected rotational diffusion rates of NATA in the presence of co-solvents demonstrates the power of joint experimental-computational approaches to biophysical and biochemical problems. We intend to apply this approach to a more complex and flexible bio-molecular system in the near future, including studies of force-field dependence of computed properties.

Acknowledgements GSJ would like to thank C.K. Johnson for allowing access to the photon counting apparatus. The computational resources used in this project were supported in part by the University of Kansas General Research Fund and by University of Kansas Research Investment Council Level II grant #36 to KK. GSJ would like to thank Prof. Zilka Rios of UCC for helpful comments.

Supplementary Information The full details of structural and dynamic properties of all studied NATA conformers in the different environments are presented in Supplementary Information.


Page 18 of 39

Page 19 of 39

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. Tanford, C. Protein denaturation. C. Theoretical models for the mechanism of denaturation. Adv. Protein Chem. 1970, 24,1 −95. 2. Nozaki, Y.; Tanford, C. The Solubility of Amino Acids, Diglycine, and Triglycine in Aqueous Guanidine Hydrochloride Solutions. J. Biol. Chem. 1970, 245, 1648−1652. 3. 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, 67− 97. 4. 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. 5. Auton, M.; Holthauzen, L. M. F.; Bolen, D. W. Anatomy of energetic changes accompanying urea-induced protein denaturation. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 15317−15322. 6. 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. 7. Canchi, D. R.; Paschek, D.; García, A. Equilibrium study of protein denaturation by urea. J. Am. Chem. Soc. 2010, 132, 2338−2344. 8. England, J. L.; Haran, G. Role of solvation effects in protein denaturation: From thermodynamics to single molecules and back. Annu. Rev. Phys. Chem. 2011, 62, 257− 277. 9. Tran, M.T.; Mao, A; Pappu, R.V. Role of backbone-solvent interactions in determining conformational equilibria of intrinsically disordered polypeptides. J. Am. Chem. Soc. 2008, 130, 7381−7392. 10. Baskakov, I.; Bolen, D. W. Forcing thermodynamically unfolded proteins to fold. J. Biol. Chem. 1998, 273, 4831−4834. 11. Qu, Y. X.; Bolen, C. L.; Bolen, D. W. Osmolyte-driven contraction of a random coil protein. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 9268−9273. 12. Wang, A.; Bolen, D. W. A naturally occurring protective system in urea-rich cells: Mechanism of osmolyte protection of proteins against urea denaturation. Biochemistry 1997, 36, 9101−9108. 13. Lin, T. Y.; Timasheff, S. N. Why do some organisms use a urea-methylamine mixture as osmolyte? Thermodynamic compensation of urea and trimethylamine N-oxide interactions with protein. Biochemistry 1994, 33, 12695− 12701.



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. Krywka, C.; Sternemann, C.; Paulus, M.; Tolan, M.; Royer, C.; Winter, R. Effect of osmolytes on pressure-induced unfolding of proteins: A high-pressure SAXS study. Chem Phys Chem 2008, 9, 2809−2815. 15. 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, 427–28. 16. Shimizu, S; Chan, H.S. Origins of protein denatured state compactness and hydrophobic clustering inaqueous urea:inferences from nonpolar potentials of mean force. Proteins Struct. Funct. Genet. 2002, 49, 560– 66. 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. 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, 7346–53. 19. England, J.L.; Pande, V.S.; Haran, G. Chemical denaturants inhibit the onset of dewetting. J. Am. Chem. Soc. 2008, 130, 11854–55. 20. Zangi, R.; Zhou, R.H.; Berne, B.J. Urea’s action on hydrophobic interactions. J. Am. Chem. Soc. 2009, 131, 1535–41. 21. 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. 22. 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. 23. Jas, G.S.; Rentchler, E.C.; Slowicka, A.M.; Hermansen, J.R.; Johnson, C.K.; Middaugh, C.R.; Kuczera, K. Reorientation motion and preferential interactions of a peptide in denaturants and osmolyte. J. Phys. Chem. B. 2016, 120, 3089−3099. 24. Cantor, C.R.; Schimmel, P.R. Biophysical Chemistry. Part II. Techniques for the Study of Biological Structure and Function; W.H. Freeman & Co: San Francisco, 1980. 25. Jas, G.S, Wang, Y., Pauls, S.W, Johnson, C.K. and Kuczera, K. Influence of temperature and viscosity on anthracene rotational diffusion in organic solvents: Molecular dynamics simulations and fluorescence anisotropy study. J. Chem. Phys. 1997, 107, 8800-8812. 26. Perrin, F. Mouvement Brownien d’un ellipsoide (II). Rotation libre et depolarisation des fluorescences. Translation et diffusion de molecules ellipsoidales, J. Chim. Radium 1936, 1, 111.


Page 20 of 39

Page 21 of 39

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. Ortega, A., Amorós, D., Garcia de la Torre, J. Prediction of hydrodynamic and other solution properties of rigid proteins from atomic- and residue-level models Biophys. J., 2011, 101, 892-898. 28. 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. 29. 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. 30. Shen, X.; Knutson, J. R. Subpicosecond fluorescence spectra of tryptophan in water. J. Phys. Chem. B 2001, 105, 6260−6265. 31. Weerasighe, S. and Smith, P. E. A Kirkwood-Buff derived force field for the simulation of aqueous guanidinium chloride solutions. J. Chem. Phys. 2004, 121, 1768938. 32. 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. 33. Karpen, M. E.; Tobias, D. T.; Brooks, C. L., III Statistical clustering techniques for the analysis of long molecular dynamics trajectories: analysis of 2.2-ns trajectories of YPGDV. Biochemistry 1993, 32, 412−420. 34. 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. 35. 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. 36. Hegefeld, W.A.; Chen, S.-E.; DeLeon, K.Y.; Kuczera, K.; Jas, G.S. Helix formation in a pentapeptide: Experimental and force-field dependent dynamics. J. Phys. Chem. A 2010, 114, 12391-12402. 37. Jas, G.S.; Kuczera, K. Computer simulations of helix folding in homo- and heteropeptides. Mol. Sim. 2012, 38, 682-694. 38. Jas, G.S.; Middaugh, C.R; Kuczera, K. Non-exponential kinetics and a complete folding pathway of an α-helical heteropeptide: Direct observation and comprehensive molecular dynamics. J. Phys. Chem. B 2014, 118, 639-647. 39. 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. 40. 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.



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

41. Smiatek, J. Osmolyte effects: impact on the aqueous solution around charged and neutral spheres. J. Phys. Chem. B 2014, 118, 771− 782. 42. 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. 43. Talis, P.G.; Melissas, V.S.; Troganis, A.N. A “hidden” role of amino and imino groups is unveiled during micro-solvation study of three biomolecule groups in water. New J. Chem. 2012, 36, 1866-1878.


Page 22 of 39

Page 23 of 39

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 Captions. Figure 1. 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 5 M urea solution; (C) green, 5 M GdmCl; (D) Orange, 5 M proline solution. (φ,ψ) and (χ1,χ2) distribution plots of NATA in the form of potential of mean force, V(x,y)=−RT ln P(x,y), where R is the gas constant, T = 300 K, and P is the probability. Coloring: red, lowest free energy; purple, highest free energy. Backbone (φ,ψ) distribution in water (A), in urea (B), in Gdmcl (C), and in proline (D). Side chain (χ1,χ2) distribution in water (E), in urea (F), in GdmCl (G), and in proline (H). Central structures of structural clusters corresponding to the main free energy minima are shown. Figure 2. Free energy as a function of conformers. (A) free energy as a function of conformers present in the aqueous solution. (B) NATA conformers in Urea, in figure (C), NATA conformers in GdmCl, and (D) NATA conformers in proline. Figure 3. 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 (280, 300, and 320 K) for urea and GdmCl and three different concentrations (0, 3.5, and 5.5 M) and three different temperatures (280, 300, and 320 K) for proline. Figure 4. Fluorescence lifetime and fluorescence decays for NATA in aqueous, urea, GdmCl, and proline solution. Fluorescence lifetime are shown in (A-D). Fluorescence decay polarized parallel to the excitation (blue), perpendicular to the excitation (green), and the instrument function (black) 0 M (E) and 5.5 M (urea F, GdmCl G, and proline, H) at 300 K. Since decays of fluorescence with polarization parallel and perpendicular to the excitation polarization were recorded simultaneously in two different detection channels with different detection efficiencies, a parameter incorporated into the fitting equations. The correction parameter had a value g ≈ 0.3 where g is the relative detection efficiency of fluorescence with polarization perpendicular to the excitation polarization relative to fluorescence with polarization parallel to the excitation polarization. Fit to the parallel and perpendicular fluorescence decays are in red. Weighted residuals are for the fits to the parallel (blue) and perpendicular (green) fluorescence decays.



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 24 of 39

Figure 5. The fluorescence anisotropy decay (A-D). Orientation times from fluorescence anisotropy measurements in two different pHs (E-F). Figure 6. C2(t) correlation function from explicit cosolvent molecular dynamics simulations of NATA illustrating rotational orientation time in 0 and 5 M urea, GdmCl and proline at 300 K (AD). Orientation times of NATA conformers with lowest free energy (∆G = 0) compared to experimental results in pH 7.2 at 300K in aqueous, 5M urea, 5M GdmCl, and 5M proline solution (E). Comparison of orientation times of NATA conformer with next to lowest ∆G and experimental measurements in pH 4.8 at 300K in aqueous, urea, GdmCl (F). Structural shapes of NATA conformers, with lowest and next to lowest free energy in four different solvent environments (G-H). Figure 7. Rotational correlation time τrot as a function of radius of gyration (Rg in Å) in (A) aqueous, (B) 5 M urea, (C) 5 M GdmCl, and (D) 5 M proline solutions. Figure 8. Translational diffusion coefficients Dt as a function of radius of gyration (Rg in Å) in (A) aqueous, (B) 5 M urea, (C) 5 M GdmCl, and (D) 5 M proline solutions. Figure 9. Rotational orientation time τrot as a function of molecular shape (δ*) in (A) aqueous, (B) 5 M urea, (C) 5 M GdmCl, and (D) 5 M proline solutions. Figure 10. Translational diffusion coefficients Dt as a function of molecular shape (δ*) in (A) aqueous, (B) 5 M urea, (C) 5 M GdmCl, and (D) 5 M proline solutions. Figure 11. Rotational orientation time τrot as a function of X/Z (molecular principal axis ratio) in (A) aqueous, (B) 5 M urea, (C) 5 M GdmCl, and (D) 5 M proline solutions. Figure 12. Translational diffusion coefficients Dt as a function of X/Z (molecular principal axis ratio) in (A) aqueous, (B) 5 M urea, (C) 5 M GdmCl, and (D) 5 M proline solutions. Figure 13. Rotational orientation τrot time as a function of solvation parameters in (A) aqueous, (B) 5 M urea, (C) 5 M GdmCl, and (D) 5 M proline solutions. Figure 14. Translational diffusion coefficients Dt as a function of solvation parameters in (A) aqueous, (B) 5 M urea, (C) 5 M GdmCl, and (D) 5 M proline solutions.


Page 25 of 39

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 15. Molecular shape (δ*) vs. solvation parameters (ACEG) and radius of gyration vs. solvation parameters (BDFH) in aqueous, 5 M urea, 5 M GdmCl, and 5 M proline solutions.



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

Figures Figure 1.


Page 26 of 39

Page 27 of 39

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.

Figure 4.


Page 28 of 39

Page 29 of 39

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.

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 39

Page 31 of 39

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 39

Page 33 of 39

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 39

Page 35 of 39

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 12.



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 13.


Page 36 of 39

Page 37 of 39

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 14.



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 15


Page 38 of 39

Page 39 of 39

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 Figure


Probing Selection Mechanism of the Most Favorable Conformation of

Recommend Documents