Vanishing amplitude of backbone dynamics causes a true protein dynamical transition: 2H NMR studies on perdeuterated C-phycocyanin

Slow global motions in biosolids studied by the deuteron stimulated echo NMR experiment

Recent 15N R1ρ-relaxation studies have shown that proteins in the solid state undergo slow, low amplitude global motion in the sub-millisecond time range. This range is at the edge of the time window for R1ρ experiments and, therefore, the motional parameters obtained by this method are not precise or reliable. In this paper, we present a 2H stimulated echo study of this type of molecular dynamics. The 2H stimulated echo experiments on a static sample allow for direct measurement of the correlation function in the time range of 10-6-10-1 s, making them well suited to study this type of molecular mobility. We have conducted a detailed analytical and numerical comparison of the correlation functions obtained from the relaxation and stimulated echo experiments, which are generally different. We have identified conditions and algorithms that enable a direct comparison of the relaxation and stimulated echo experimental results. Using the protein GB1 in the form of a lyophilized powder, we have demonstrated that 15N R1ρ-relaxation and 2H stimulated echo experiments yield essentially the same slow-motion correlation function. Surprisingly, this type of motion is observed not only in the protein sample but also in the tripeptide and single amino acid solid samples. The comparison of data measured in these three samples at different temperatures led us to conclude that this slow motion is, in fact, ultrasonic phonons, which seem to be inherent to all rigid biological solids.

Temperature-Dependent Dynamics at Protein-Solvent Interfaces

We perform differential scanning calorimetry, broadband dielectric spectroscopy (BDS), and nuclear magnetic resonance (NMR) studies to ascertain the molecular dynamics in mixtures of ethylene glycol with elastin or lysozyme over broad temperature ranges. To focus on the protein-solvent interface, we use mixtures with about equal numbers of amino acids and solvent molecules. The elastin and lysozyme mixtures show similar glass transition steps, which extend over a broad temperature range of 157-185K. The BDS and NMR studies yield fully consistent results for the fastest process P1, which is caused by the structural relaxation of ethylene glycol between the protein molecules and follows an Arrhenius law with an activation energy of Ea=0.63eV. It involves quasi-isotropic reorientation and is very similar in the elastin and lysozyme matrices but different from the alpha and beta relaxations of bulk ethylene glycol. Two slower BDS processes P2 and P3 have protein-dependent time scales, but exhibit a similar Arrhenius-like temperature dependence with an activation energy of Ea~0.81eV. However, P2 and P3 do not have a clear NMR signature. In particular, the NMR results for the lysozyme mixture reveal that the protein backbone does not show isotropic alpha-like motion on the P2 and P3 time scales but only restricted beta-like reorientation. The different activation energies of the P1 and P2/P3 processes do not support an intimate coupling of protein and ethylene glycol dynamics. The present results are compared with previous findings for mixtures of proteins with water or glycerol, implying qualitatively different dynamical couplings at various protein-solvent interfaces.

Structural and Dynamical Properties of Liquids in Confinements: A Review of Molecular Dynamics Simulation Studies

Molecular dynamics (MD) simulations are a powerful tool for detailed studies of altered properties of liquids in confinement, in particular, of changed structures and dynamics. They allow, on one hand, for perfect control and systematic variation of the geometries and interactions inherent in confinement situations and, on the other hand, for type-selective and position-resolved analyses of a huge variety of structural and dynamical parameters. Here, we review MD simulation studies on various types of liquids and confinements. The main focus is confined aqueous systems, but also ionic liquids and polymer and silica melts are discussed. Results for confinements featuring different interactions, sizes, shapes, and rigidity will be presented. Special attention will be given to situations in which the confined liquid and the confining matrix consist of the same type of particles and, hence, disparate liquid-matrix interactions are absent. Findings for the magnitude and the range of wall effects on molecular positions and orientations and on molecular dynamics, including vibrational motion and structural relaxation, are reviewed. Moreover, their dependence on the parameters of the confinement and their relevance to theoretical approaches to the glass transition are addressed.

Confinement Effects on Glass-Forming Aqueous Dimethyl Sulfoxide Solutions

Combining broadband dielectric spectroscopy and nuclear magnetic resonance studies, we analyze the reorientation dynamics and the translational diffusion associated with the glassy slowdown of the eutectic aqueous dimethyl sulfoxide solution in nano-sized confinements, explicitly, in silica pores with different diameters and in ficoll and lysozyme matrices at different concentrations. We observe that both rotational and diffusive dynamics are slower and more heterogeneous in the confinements than in the bulk but the degree of these effects depends on the properties of the confinement and differs for the components of the solution. For the hard and the soft matrices, the slowdown and the heterogeneity become more prominent when the size of the confinement is reduced. In addition, the dynamics are more retarded for dimethyl sulfoxide than for water, implying specific guest-host interactions. Moreover, we find that the temperature dependence of the reorientation dynamics and of the translational diffusion differs in severe confinements, indicating a breakdown of the Stokes-Einstein-Debye relation. It is discussed to what extent these confinement effects can be rationalized in the framework of core-shell models, which assume bulk-like and slowed-down motions in central and interfacial confinement regions, respectively.

Quasielastic neutron scattering studies on couplings of protein and water dynamics in hydrated elastin

Performing quasielastic neutron scattering measurements and analyzing both elastic and quasielasic contributions, we study protein and water dynamics of hydrated elastin. At low temperatures, hydration-independent methyl group rotation dominates the findings. It is characterized by a Gaussian distribution of activation energies centered at about E_m = 0.17 eV. At ∼195 K, coupled protein-water motion sets in. The hydration water shows diffusive motion, which is described by a Gaussian distribution of activation energies with E_m = 0.57 eV. This Arrhenius behavior of water diffusion is consistent with previous results for water reorientation, but at variance with a fragile-to-strong crossover at ∼225 K. The hydration-related elastin backbone motion is localized and can be attributed to the cage rattling motion. We speculate that its onset at ∼195 K is related to a secondary glass transition, which occurs when a β relaxation of the protein has a correlation time of τ_β ∼ 100 s. Moreover, we show that its temperature-dependent amplitude has a crossover at the regular glass transition T_g = 320 K of hydrated elastin, where the α relaxation of the protein obeys τ_α ∼ 100 s. By contrast, we do not observe a protein dynamical transition when water dynamics enters the experimental time window at ∼240 K.

On the microscopic origins of relaxation processes in aqueous peptide solutions undergoing a glass transition

We combine broadband dielectric spectroscopy (BDS) with ¹H and ²H nuclear magnetic resonance (NMR) to study molecular dynamics in mixtures of ε-polylysine with H₂O or D₂O. In BDS, four relaxation processes can be attributed to molecular dynamics. While the fastest process P1 obeys the Arrhenius law, the slowest process P4 shows prominent non-Arrhenius behavior typical of structural α relaxation. For the intermediate processes P2 and P3, the temperature dependence changes at the glass transition temperature T_g. The ¹H and ²H NMR results yield insights into the molecular origins of these relaxation phenomena. In these NMR analyses, we exploit, in addition to the isotope selectivity of the method, the possibility to distinguish between various types of motion based on their respective line-shape effects and the capability to single out specific molecular moieties based on different spin-lattice relaxation behaviors. In this way, we reveal that process P1 results from the rotation of side and end groups of the peptide, while process P2 is caused by a reorientation of essentially all water molecules, which are quasi-isotropic and survive well below T_g. As for the peptide backbone dynamics, we find evidence that rotational motion of polar groups is involved in process P3 and that nonpolar regions show a dynamical process, which is located between P3 and P4. Thus, the NMR analyses do not yield evidence for coexisting fast peptide-decoupled and slow peptide-coupled water species, which contribute to BDS processes P2 and P3, respectively, but minor bimodality of water motion may remain undetected. Finally, it is demonstrated that the proton/deuteron exchange needs to be considered when interpreting experimental results for molecular dynamics in aqueous peptide solutions.

Redox chemistry in the pigment eumelanin as a function of temperature using broadband dielectric spectroscopy

We demonstrate on synthetic eumelanin that biomolecular conductivity models should account for temperature and hydration effects coherently.

2H NMR Studies on the Dynamics of Pure and Mixed Hydrogen-Bonded Liquids in Confinement

2H NMR is used to ascertain dynamical behaviors of pure and mixed hydrogen-bonded liquids in bulk and in confinement. Detailed comparisons of previous and new results in broad dynamic and temperature ranges reveal that confinement effects differ for various liquids and confinements. For water, molecular reorientation strongly depends on the confinement size, with much slower and less fragile structural relaxation under more severe geometrical restriction. Moreover, a dynamical crossover occurs when a fraction of solid water forms so that the dynamics of the fraction of liquid water becomes even more restricted and, as a consequence, changes from bulk-like to interface-dominated. For glycerol, by contrast, confinement has weak effects on the reorientation dynamics. Mixed hydrogen-bonded liquids show even more complex dynamical behaviors. For aqueous solutions, the temperature dependence of the structural relaxation becomes discontinuous when the concentration changes due to a freezing of water fractions. This tendency for partial crystallization is enhanced rather than reduced by confinement, because different liquid-matrix interactions of the molecular species induce micro-phase segregation, which facilitates ice formation in water-rich regions. In addition, dynamical couplings at solvent-protein interfaces are discussed. It is shown that, on the one hand, solvent dynamics are substantially slowed down at protein surfaces and, on the other hand, protein dynamics significantly depend on the composition and, thus, the viscosity of the solvent. Furthermore, a protein dynamical transition occurs when the amplitude of water-coupled restricted backbone dynamics vanishes upon cooling.

Fast Motions of Key Methyl Groups in Amyloid-β Fibrils

Amyloid-β (Aβ) peptide is the major component of plaques found in Alzheimer's disease patients. Using solid-state 2H NMR relaxation performed on selectively deuterated methyl groups, we probed the dynamics in the threefold symmetric and twofold symmetric polymorphs of native Aβ as well as the protofibrils of the D23N mutant. Specifically, we investigated the methyl groups of two leucine residues that belong to the hydrophobic core (L17 and L34) as well as M35 residues belonging to the hydrophobic interface between the cross-β subunits, which has been previously found to be water-accessible. Relaxation measurements performed over 310-140 K and two magnetic field strengths provide insights into conformational variability within and between polymorphs. Core packing variations within a single polymorph are similar to what is observed for globular proteins for the core residues, whereas M35 exhibits a larger degree of variability. M35 site is also shown to undergo a solvent-dependent dynamical transition in which slower amplitude motions of methyl axes are activated at high temperature. The motions, modeled as a diffusion of methyl axis, have activation energy by a factor of 2.7 larger in the twofold compared with the threefold polymorph, whereas D23N protofibrils display a value similar to the threefold polymorph. This suggests enhanced flexibility of the hydrophobic interface in the threefold polymorph. This difference is only observed in the hydrated state and is absent in the dry fibrils, highlighting the role of solvent at the cavity. In contrast, the dynamic behavior of the core is hydration-independent.

Dynamics of Hydrophobic Core Phenylalanine Residues Probed by Solid-State Deuteron NMR

We conducted a detailed investigation of the dynamics of two phenylalanine side chains in the hydrophobic core of the villin headpiece subdomain protein (HP36) in the hydrated powder state over the 298-80 K temperature range. Our main tools were static deuteron NMR measurements of longitudinal relaxation and line shapes supplemented with computational modeling. The temperature dependence of the relaxation times reveals the presence of two main mechanisms that can be attributed to the ring-flips, dominating at high temperatures, and small-angle fluctuations, dominating at low temperatures. The relaxation is nonexponential at all temperatures with the extent of nonexponentiality increasing from higher to lower temperatures. This behavior suggests a distribution of conformers with unique values of activation energies. The central values of the activation energies for the ring-flipping motions are among the smallest reported for aromatic residues in peptides and proteins and point to a very mobile hydrophobic core. The analysis of the widths of the distributions, in combination with the earlier results on the dynamics of flanking methyl groups (Vugmeyster et al. J. Phys. Chem. B 2013, 117, 6129-6137), suggests that the hydrophobic core undergoes slow concerted fluctuations. There is a pronounced effect of dehydration on the ring-flipping motions, which shifts the distribution toward more rigid conformers. The crossover temperature between the regions of dominance of the small-angle fluctuations and ring-flips shifts from 195 K in the hydrated protein to 278 K in the dry one. This result points to the role of solvent in softening the core and highlights aromatic residues as markers of the protein dynamical transitions.

Effects of solvent concentration and composition on protein dynamics: 13C MAS NMR studies of elastin in glycerol-water mixtures

We use (13)C CP MAS NMR to investigate the dependence of elastin dynamics on the concentration and composition of the solvent at various temperatures. For elastin in pure glycerol, line-shape analysis shows that larger-scale fluctuations of the protein backbone require a minimum glycerol concentration of ~0.6 g/g at ambient temperature, while smaller-scale fluctuations are activated at lower solvation levels of ~0.2 g/g. Immersing elastin in various glycerol-water mixtures, we observe at room temperature that the protein mobility is higher for lower glycerol fractions in the solvent and, thus, lower solvent viscosity. When decreasing the temperature, the elastin spectra approach the line shape for the rigid protein at 245 K for all studied samples, indicating that the protein ceases to be mobile on the experimental time scale of ~10(-5) s. Our findings yield evidence for a strong coupling between elastin fluctuations and solvent dynamics and, hence, such interaction is not restricted to the case of protein-water mixtures. Spectral resolution of different carbon species reveals that the protein-solvent couplings can, however, be different for side chain and backbone units. We discuss these results against the background of the slaving model for protein dynamics.