Effect of conformational states on protein dynamical transition

https://doi.org/10.1016/j.bbapap.2009.06.025Get rights and content

Abstract

In order to examine the properties specific to the folded protein, the effect of the conformational states on protein dynamical transition was studied by incoherent elastic neutron scattering for both wild type and a deletion mutant of staphylococcal nuclease. The deletion mutant of SNase which lacks C-terminal 13 residues takes a compact denatured structure, and can be regarded as a model of intrinsic unstructured protein. Incoherent elastic neutron scattering experiments were carried out at various temperature between 10 K and 300 K on IN10 and IN13 installed at ILL. Temperature dependence of mean-square displacements was obtained by the q-dependence of elastic scattering intensity. The measurements were performed on dried and hydrated powder samples. No significant differences were observed between wild type and the mutant for the hydrated samples, while significant differences were observed for the dried samples. A dynamical transition at ∼ 140 K observed for both dried and hydrated samples. The slopes of the temperature dependence of MSD before transition and after transition are different between wild type and the mutant, indicating the folding induces hardening. The hydration water activates a further transition at ∼ 240 K. The behavior of the temperature dependence of MSD is indistinguishable for wild type and the mutant, indicating that hydration water dynamics dominate the dynamical properties.

Introduction

It is widely accepted that the unique tertiary structure of a protein is required for its specific function. The structure–function relationship is thus a central subject in protein science even in the post-genomic era. Many studies have demonstrated that the internal protein motion is required for the function [1], [2], [3], [4], [5]. Therefore, there arises the question whether a protein acquires the specific dynamical properties upon folding. The dynamical differences between a native and non-native proteins have been characterized [6], [7], [8], [9], [10], [11], [12]. These studies suggest that the high-frequency, bond-vibrational and angle-bending motions are almost identical regardless of their conformations [13], [14], whereas the lower-frequency collective dynamics are different. An additional factor to be considered is the solvent. Although the potential surface of a protein in vacuo is intrinsically anharmonic [15], it is also accepted that solvent induces the dynamical transition from harmonic to anharmonic dynamical behavior in globular proteins i.e., protein dynamics are strongly coupled to their solvent environment [16], [17], [18], [19], [20], [21], [22], [23].

Incoherent neutron scattering (INS) is useful and effective for the studies of the internal dynamics of proteins in the picosecond–nanosecond time scale [24]. INS probes the self time-correlations of the hydrogen atom positions sensitively due to its large incoherent cross section. Because hydrogen atoms are evenly distributed throughout a protein, this technique gives a global view of the protein motions. The dependence of elastic incoherent scattering intensity on the scattering vector (q) provides the average mean-square displacements (MSD) in proteins [17], [25]. Derivation of the MSD requires the assumption that the elastic scattering is approximated by a Gaussian function [23], [26]. It is usually valid the Gaussian approximation in low q region, but in wide q range, the deviation from a Gaussian becomes remarkable [27], [28]. The non-Gaussian behavior can occur from the individual atoms exhibiting the non-Gaussian dynamics (such as jump diffusion) and/or by the existence of more than one MSD value in the proteins, called as the dynamical heterogeneity [15], [29]. We reported that the dynamical heterogeneity dominantly contributes to the non-Gaussian behavior [29], [30], [31]. The dynamical heterogeneity leads to the non-Gaussian scattering even when the individual atom scattering is a Gaussian [15], [32]. We have also shown that a bimodal distribution is adequate for the quantitative analysis of the experimentally observed scattering function in terms of dynamical heterogeneity, by which the protein internal motions can be separated into the inner core and surface part in the protein structure [29].

The time scale of the motions observed by INS is determined by the instrumental energy resolution [33], [34]. Protein motions can be observed when the time scale of the motions is within the instrumental frequency windows. Daniel et al. showed experimentally that the observed dynamics of the enzyme, glutamate dehydrogenase (GDH), possesses a marked dependence on the time scale [35]. The molecular dynamics simulations demonstrated that the observed dynamical transition is varied by the instrumental energy resolution [33], [35].

We focus here on the comparison of dynamics between a folded and partially unfolded protein, Staphylococcal nuclease (SNase). The partial unfolding was achieved by the truncation of 13C-terminal residues [36], [37], [38], [39], [40]. The resulting structure is well described as a hydrophobic core with flaring tails [37]. To characterize the dynamical differences between the two conformational states, the incoherent elastic neutron scattering experiments in a wide q range and with the several instrumental energy resolutions were performed. The non-Gaussian behavior in wide q-range was analyzed by the bimodal distribution model. The fraction of the surface flexible part increases by partial unfolding. The difference in dynamical heterogeneity between two different conformational states appeared more distinctly in the dried condition. In contrast, the hydration makes this difference obscure, indicating that the protein dynamics is influenced by hydration. We also found that the apparent dynamical transition depended on the experimentally observable time scale and the time scale dependent dynamics were different between the wild type and the truncated mutant for both the dried and hydrated samples. The hierarchy of the protein dynamics in the time scale is changed by protein folding, and the native and non-native proteins have the different dynamical structures.

Section snippets

Sample preparation

Wild type and truncated SNase were expressed in Escherichia coli. The proteins were purified by urea extraction and with SP-Sepharose Fast Flow column chromatography. The purified proteins were dialyzed against water and lyophilized. The lyophilized proteins were dissolved in D2O for the H-D exchange of the labile hydrogen atoms, and lyophilized again. This procedure was repeated two or three times. The lyophilized proteins thus obtained were used as the dried samples. In order to prepare the

Structure

To control the protein hydration level, the lyophilized samples were used in the INS experiments. It is important to determine if there are unsuitable structural changes by lyophilization. We measured the CD spectra of the proteins with lyophilized film, and compared them to the corresponding spectra in solution. Fig. 1A shows the CD spectra of the wild type and the truncated mutant in solution. The negative peak at 205 nm arises from the random coil structure, and the peak at 222 nm indicates

Instrumental energy resolution and apparent transition

The finite energy resolution effect can be recognized by separating the observed MSD into a converged 〈u2conv and a resolution-dependent contribution 〈u2Res in the Eq. (7) [31], [32], which brings the two possible origins of apparent (observed) dynamical transition behavior. One is described by the transition in terms of a change in the (long time) equilibrium position distribution of the protein conformation, which describes 〈u2EISF in Eq. (7). The transition temperature of this origin is

Conclusion

Conformation dependent dynamical properties were examined using the wild type folded SNase and the partially unfolded mutant. Clear differences were observed for the dried state. The force constant of the mutant is smaller than that of wild type at low temperature range. The magnitude of MSD above 150 K for the mutant is systematically larger than that of wild type. The dynamical heterogeneity is also different in the dried state. These observations support that the partially unfolded state is

References (46)

  • A. Paciaroni et al.

    Effect of the environment on the protein dynamical transition: a neutron scattering study

    Biophys. J.

    (2002)
  • A. Tournier et al.

    Translational water dynamics drives the protein dynamical transition

    Biophys. J.

    (2003)
  • W. Doster et al.

    Protein–water displacement distributions

    Biochim. Biphys. Acta

    (2005)
  • U. Lehnert et al.

    Thermal motions in bacteriorhodopsin at different hydration levels studied by neutron scattering: correlation with kinetics and light-induced conformational changes

    Biophys. J.

    (1998)
  • H. Nakagawa et al.

    Dynamical heterogeneity of protein dynamics studied by elastic incoherent neutron scattering and molecular simulations

    Mater. Sci. Eng. A.

    (2006)
  • J.A. Hayward et al.

    Temperature dependence of protein dynamics: computer simulation analysis of neutron scattering properties

    Biophys. J.

    (2002)
  • T. Becker et al.

    Neutron frequency windows and the protein dynamical transition

    Biophys. J.

    (2004)
  • R.M. Daniel et al.

    Enzyme activity below the dynamical transition at 220 K

    Biophys. J.

    (1998)
  • H. Nakagawa et al.

    Hydration affects both harmonic and anharmonic nature of protein dynamics

    Biophys. J.

    (2008)
  • V. Reat et al.

    Dynamics of different functional parts of bacteriorhodopsin: H–2H labeling and neutron scattering

    Proc. Natl. Acad. Sci. U. S. A.

    (1998)
  • A. Ostermann et al.

    Ligand binding and conformational motions in myoglobin

    Nature

    (2000)
  • B.J. Rasmussen et al.

    Crystalline ribonuclease A loses function below the dynamical transition at 220 K

    Nature

    (1992)
  • J.C. Smith et al.

    Motions in native and denatured proteins

    Physica B

    (1998)
  • Cited by (22)

    • Rigidity of protein structure revealed by incoherent neutron scattering

      2020, Biochimica et Biophysica Acta - General Subjects
      Citation Excerpt :

      The dynamical differences between folded and unfolded proteins were also confirmed in solution at room temperature [55]. A difference in dynamical heterogeneity was observed between the folded and partially unfolded SNase, and the fraction of rigid internal core in the protein's structure decreased upon partial unfolding [56]. This result demonstrates that the rigidity specific to the folded protein is reflected in its dynamical heterogeneity as it relates to atomic packing; however, the question arises as to whether the Boson peak comes from the modes based on secondary structural elements, or from those extended over the whole molecule.

    • Universality and Structural Implications of the Boson Peak in Proteins

      2019, Biophysical Journal
      Citation Excerpt :

      SNase is one of the typical globular proteins as shown in Fig.1, and its molecular weight is ∼1.7 kDa. This protein has been used as a model protein for studying protein dynamics by inelastic neutron scattering (18,19,30,37,39,45,52). The physical properties of SNase are similar to the other globular proteins, such as myoglobin, lysozyme, and so on.

    • Quasielastic neutron scattering in biology: Theory and applications

      2017, Biochimica et Biophysica Acta - General Subjects
      Citation Excerpt :

      According to the modeling implicit in Fig. 10, the existence of the dynamical transitions is not an artificial consequence of finite instrumental resolution W and a correspondingly limited time window. However, the instrumental energy resolution does have a significant effect on the theoretical TD, shifting it to higher temperature with increasing energy resolution width [73–75,78,79]. The impact of the energy resolution width (W) on the MSD is determined by the decay rate of the correlations, which depends on the relaxation frequency parameter λ (1/τ) in the stretched exponential function C(t).

    • Picosecond dynamics in haemoglobin from different species: A quasielastic neutron scattering study

      2014, Biochimica et Biophysica Acta - General Subjects
      Citation Excerpt :

      It is, however, in general completely unrealistic to assume that all dynamic processes in proteins are characterized only by a single relaxation rate. Mean square displacements, for example, measured by neutron scattering depend very strongly on the instrumental energy resolution [50,51]. The present study concerning fast picosecond dynamics in Hb is a continuation of previous work, where we investigated the dynamics of Hb from different species on the slower 0.1 ns time scale [13].

    • Dynamical coupling of intrinsically disordered proteins and their hydration water: Comparison with folded soluble and membrane proteins

      2012, Biophysical Journal
      Citation Excerpt :

      Interestingly, unfolded tRNA showed smaller MSDs than folded tRNA (49), in contrast to the opposite behavior of proteins. Even in the absence of hydration water, tau is characterized at room temperature by larger MSDs than are dry folded proteins (Fig. 3), in line with results obtained on dry powders of folded wild-type and partially unfolded mutant staphylococcal nuclease (47). Whether disordered (Fig. 1 A) or denatured, proteins thus display a much broader dynamical spectrum than do their folded counterparts (Fig. 2).

    View all citing articles on Scopus
    View full text