Conformation of tetranactin in solution.

The conformation of tetranactin, an ionophore, in chloroform was investigated by infrared and Raman spectra and by proton and ¹³C magnetic resonances. The infrared spectra show that the structure of its K⁺ complex in the solution is quite similar to that in crystals. The proton spin–spin coupling constants are explained well by assuming that the crystalline structure is retained in solution. The spin–lattice relaxation times of the ¹³C nuclei of the K⁺ complex indicate that its framework is rigid. The correlation time of the overall reorientation of the molecule was calculated to be 9 X 10^?11 sec. On the other hand, the conformation of the complexed form in chloroform differs from that in crystals. Despite the geometrical nonequivalence of the four subunits in the crystalline state, the nuclear magnetic resonance spectra show their magnetic equivalence in the solution. The proton spin–spin coupling constants have values that are averaged by rapid internal rotation. The spin–lattice relaxation times of the ¹³C nuclei in its framework are unexplained by the overall reorientation of the molecule, and reveal the existence of internal motion in the framework. The rate of the local motion of the framework is between 10²–10¹⁰ sec^?1. By comparison of the infrared spectra, it can be said that the mean conformation of the fluctuated framework of the uncomplexed tetranactin in the solution is similar to that of nonactin in the crystalline form, which has an S₄ symmetry axis through the center of the macrocyclic ring.     

Smoluchowski dynamics of the vnd/NK‐2 homeodomain from Drosophila melanogaster: First‐order mode‐coupling approximation

This work is the first in a series devoted to applying mode coupling diffusion theory to the derivation of local dynamics properties of proteins in solution. The first‐order mode‐coupling approximation, or optimized Rouse–Zimm local dynamics (ORZLD), is applied here to derive the rotational dynamics of the bonds and compare the calculated with the experimental nmr ¹⁵N spin–lattice relaxation time behavior of the vnd/NK‐2 homeodomain from Drosophila melanogaster. The starting point for the calculations is the experimental three‐dimensional structure of the homeodomain determined by multidimensional nmr spectroscopy. The results of the computations are compared with experimentally measured ¹⁵N spin–lattice relaxation times T₁, at 34.5 and 60.8 MHz, to check the first‐order approximation. To estimate the relative importance of internal and overall rotation, both rigid and fluctuating dynamic models are examined, with fluctuations evaluated using molecular dynamics (MD) simulations. The correlation times for the fundamental bond vector time correlation function and for the second‐order bond orientational TCF are obtained as a function of the residue number for vnd/NK‐2. The stability of the corresponding local dynamics pattern for the fluctuating structure as a function of the length of the MD trajectory is presented. Diffusive dynamics, which is essentially free of model parameters even at first order in the mode‐coupling diffusion approach, confirm that local dynamics of proteins can be described in terms of rotational diffusion of a fluctuating quasi‐rigid structure. The comparison with the nmr data shows that the first‐order mode coupling diffusion approximation accounts for the correct order of magnitude of the results and of important qualitative aspects of the data sensitive to conformational changes. Indications are obtained from this study to efficiently extend the theory to higher order in the mode‐coupling expansion. These results demonstrate the promise of the mode‐coupling approach, where the local dynamics of proteins is described in terms of rotational diffusion of a fluctuating quasi‐rigid structure, to analyze nmr spin–lattice relaxation behavior. © 1999 John Wiley & Sons, Inc. Biopoly 49: 235–254, 1999     

NMR field-cycling relaxation spectroscopy of bovine serum albumin,muscle tissue,micrococcus luteus and yeast: 14N1H-quadrupole dips

The frequency dependence of the proton spin lattice relaxation time of bovine serum albumin, muscle tissue, Micrococcus luteus and yeast has been measured by the aid of the field-cycling technique. In all systems ¹⁴N¹H-quadrupole dips have been observed. The conclusion is that amide groups are the dominating relaxation centers up to approx. 10⁷Hz. This finding can be understood by the fact that protein backbone fluctuations and, if possible, tumbling of the whole molecule rather than side group motions are the relevant mechanisms in this frequency range. A proton relaxation scheme for cells and tissue is presented.     

Nmr studies on the presence of domain structures in polysaccharides

A variety of nmr transient techniques demonstrate the absence of domain structures in powders of α- and β-cyclodextrins, dextran B512F, and their “deuterated” analogs where deuterium replaces exchangeable protons. Cyclodextrins and dextran are inferred to be homogeneously ordered and disordered one-phase systems, respectively. We show how erroneous results concerning the presence and sizes of rigid and mobile domains can be reached from the interpretations of two-component free induction decays and spin–lattice relaxation behavior of polysaccharides. In our approach, we have measured ¹H free induction decays and lineshapes, and ¹H spin–lattice and dipolar relaxation times, in addition to using Goldman–Shen experiments and ¹³C cross-polarization magic-angel sample spinning. Self-consistent nmr results are suggested as essential means to corroborate conclusions concerning domain structures in biopolymers.     

Thermal transition of DNA measured by NMR spin-echo technique

The thermal helix–coil transition of DNA can be studied by means of the spin-echo technique. The longitudinal spin–lattice relaxation time T1 and the transvense spin–spin relaxation time T2 of the DNA sample show a similar phase transition as observed spec-trophotometrically with increasing and decreasing temperatures. Four slopes on the T1 and T2 temperature relationship curves were found and interpreted as functions of nonrelational hydration of the DNA molecule. The T1 and T2 values differ depending on the native or denatured state of the DNA molecule. The importance of the dynamic equilibrium between water molecules in the hydration lattice and steps in the denaturation of the DNA molecule are discussed. This phenomenon may be directly related to the nonrotational hydration.     

Distribution of Mn2+ ions around poly(rI).poly(rC)

The distribution of bound Mn²⁺ ions about poly(rI)·poly(rC) has been studied by measuring the effect of this paramagnetic metal ion on the relaxation behavior of poly(rI)·poly(rC) protons. By combining selective spin – lattice and spin – spin relaxation rates for various protons, some of the principle regions of ion association can be identified. The relaxation data on the CH6 proton are consistent with a < 10% occupancy of phosphate inner-sphere binding sites. The broadening of the imino proton resonance requires a substantial occupancy of sites located in the major groove, possibly near IN7. This would also be consistent with the observation that IH8 resonance is the proton most susceptible to relaxation by Mn²⁺. The relaxation data for the IH2 proton indicate a relatively low occupancy of minor-groove binding sites (e.g., IN3).     

Anisotropic rotational motions of poly(L-glutamic acid) in the alpha-helix conformation

The anisotropic rotational motion of the backbone and the side chains of poly(L -glutamic acid) in the α-helical structure was investigated using the ¹³C-T₁ and T₂ relaxation times of all carbon atoms with directly attached protons, obtained at a ¹³C-Larmor frequency of 67.89 MHz. The evaluation of the nmr data was carried out according to the previously derived anisotropic diffusion model, in which the macromolecule is considered a rigid rod. The rotation of the backbone is characterized by two diffusion constants, D₁ and D₃, describing the rotation perpendicular to and around the symmetry axis. The additional internal motion of the C^β-methylene group is described as a jump process with a jump rate, k₁, between two allowed rotametric states. Steric considerations indicate that the occupation of the third rotameric position is forbidden. The rotation of the C^γ-methylene group is decribed as a one-dimensional diffusion process around the C^β–C^γ bond. Investigation of the temperature dependence of the relaxation parameters led to the temperature dependence of the dynamic parameters. Activation energies were determined from these data. The dynamic parameters obtained for poly(L -glutamic acid) at 291 K are compared with the corresponding results of a previous study of poly(L -lysine). The development of an anisotropic diffusion model for the motions of the rod-shaped poly(L -lysine) α-helix and its application to the interpretation of the ¹³C-relaxation data of this molecule have already been published previously. In this model, both the overall molecular tumbling and the various internal motions have been characterized by diffusion constants or jump rates typical for each process. These dynamic parameters can be calculated from the spin–lattice relaxation times, the spin–spin relaxation times and the NOE factors of the C^α, C^β, and C^γ nuclei of the polypetide. In the present paper, we describe the application of the above-mentioned dynamic model to the interpretation of ¹³C-relaxation studies of a further homopolypeptide, poly(L -glutamic acid), in the α-helical structure. Furthermore, we studied the temperature dependence of the relaxation times of this polymer and determined the anisotropic diffusion parameters at each temperature. From their temperature dependence and from comparison of our present results with the data of our previous study of poly(L -lysine), we were able to derive new insights into the intramolecular diffusion processes and the excitation of various motions.     

Effect of thermal denaturation on water–collagen interactions: NMR relaxation and differential scanning calorimetry analysis

The dependence of the proton spin–lattice relaxation rate, and of the enthalpy and temperature of denaturation on water content, were studied by nmr and differential scanning calorimetry (DSC) in native and denatured collagen. Collagen was first heated at four different temperatures ranging from 40 to 70°C. The percentage of denatured collagen induced by these preheating treatments was determined from DSC measurements. The DSC results are discussed in terms of heat‐induced structural changes. A two‐exponential behavior for the spin–lattice relaxation was observed with the appearance of denatured collagen. This was attributed to the presence of a noncollagen protein fraction. The variations in the different longitudinal relaxation rates as a function of the moisture content and of the denatured collagen percentage are described within the multiphase water proton exchange model. This study highlights the complementarity of the information obtained from the two analytical tools used. © 1999 John Wiley & Sons, Inc. Biopoly 50: 690–696, 1999     

A pulsed N.M.R study of D2O bound to 1,2 dipalmitoyl phosphatidylcholine

Spin lattice relaxation times in both the lab and rotating frame, have been measured for deuterons (²H) in a number of unsonicated dispersions of 1,2 dipalmitoyl phosphatidylcholine in D₂O over a range of resonant frequencies from 13 MHz to 1 MHz for temperatures from ?20°C to 65°C.The proton (¹H) spin lattice relaxation time for the lecithin was measured for resonant frequencies of 8.5 MHz, and 40 MHz over a similar range of temperatures.The results agree with broadline measurements by Salsbury et al. [1], and for the liquid crystal phase are consistent with an anisotropic tumbling model of the water molecules bound to the lecithin headgroup. This tumbling occurs with correlation times of ≤10^?10 sec and ≈ 10^?6 sec about axes parallel to and perpendicular to the bisector of the D-O-D angle within a D₂O molecule, hydrogen bonded to the negatively charged phosphate headgroup.     

Interactions of DNA with divalent metal ions. I. 31P-nmr studies

³¹P-nmr has been used to investigate the specific interaction of three divalent metal ions, Mg²⁺, Mn²⁺, and Co⁺², with the phosphate groups of DNA. Mg²⁺ is found to have no significant effect on any of the ³¹P-nmr parameters (chemical shift, line-width, T₁, T₂, and NOE) over a concentration range extending from 20 to 160 mM. The two paramagnetic ions, Mn²⁺ and Co²⁺, on the other hand, significantly change the ³¹P relaxation rates even at very low levels. From an analysis of the paramagnetic contributions to the spin–lattice and spin–spin relaxation rates, the effective internuclear metal–phosphorus distances are found to be 4.5 ± 0.5 and 4.1 ± 0.5 Å for Mn²⁺ and Co²⁺, respectively, corresponding to only 15 ± 5% of the total bound Mn²⁺ and Co²⁺ being directly coordinated to the phosphate groups (inner-sphere complexes). This result is independent of any assumptions regarding the location of the remaining metal ions which may be bound either as outer-sphere complexes relative to the phosphate groups or elsewhere on the DNA, possibly to the bases. Studies of the temperature effects on the ³¹P relaxation rates of DNA in the absence and presence of Mn²⁺ and Co²⁺ yielded kinetic and thermodynamic parameters which characterize the association and dissociation of the metal ions from the phosphate groups. A two-step model was used in the analysis of the kinetic data. The lifetimes of the inner-sphere complexes are 3 × 10^?7 and 1.4 × 10^?5 s for Mn²⁺ and Co²⁺, respectively. The rates of formation of the inner-sphere complexes with the phosphate are found to be about two orders of magnitude slower than the rate of the exchange of the water of hydration of the metal ions, suggesting that expulsion of water is not the rate-determining step in the formation of the inner-sphere complexes. Competition experiments demonstrate that the binding of Mg²⁺ ions is 3–4 times weaker than the binding of either Mn²⁺ or Co²⁺. Since the contribution from direct phosphate coordination to the total binding strength of these metal ion complexes is small (～15%), the higher binding strength of Mn²⁺ and Co²⁺ may be attributed either to base binding or to formation of stronger outer-sphere metal–phosphate complexes. At high levels of divalent metal ions, and when the metal ion concentration exceeds the DNA–phosphate concentration, the fraction of inner-sphere phosphate binding increases. In the presence of very high levels of Mg²⁺ (e.g., 3.1M), the inner-sphere ? outer-sphere equilibrium is shifted toward ～100% inner-sphere binding. A comparison of our DNA results and previous results obtained with tRNA indicates that tRNA and DNA have very similar divalent metal ion binding properties. A comparison of the present results with the predictions of polyelectrolyte theories is presented.     

High resolution proton magnetic resonance of sonicated phospholipids

We have recorded high resolution proton magnetic resonance spectra of sonicated phospholipid vesicles. The following lipids were used in separate experiments: phosphatidylglycerol, phosphatidylserine, phosphatidylethanolamine, and phosphatidylcholine from egg yolk as well as dimyristoyl phosphatidylcholine. Mixed lipid vesicles were also investigated. Assignments of the peaks associated with the various protons of the different lipids are presented. It is shown that in favorable cases, it is possible to resolve the different phospholipid head groups of mixed lipid samples. Spin lattice relaxation times (T₁) of each peak were collected at 500 MHz and 90 MHz. The influence of the addition of a small concentration of spin labeled phospholipid on i) the linewidths ii) the spin lattice relaxation times, was determined. It is shown that nitroxide radicals selectively broaden the peaks associated with the protons localized at a comparable depth of the bilayer. On the other hand, T₁ are less selectively perturbed. Potential applicability of ¹H-NMR for the investigation of lipid-proton specificity in membranes is discussed.     

Electron-electron distances in spin-labeled low-spin metmyoglobin variants by relaxation enhancement

Thirteen single-cysteine variants of myoglobin were prepared by overexpression of apoprotein, spin labeling, and reconstitution with hemin. This procedure resulted in a protein with fewer hemichrome impurities than was obtained by an overexpression of holo-protein followed by spin labeling. Coordination of cyanide to the met heme formed low-spin complexes. Iron-nitroxyl interspin distances in the range of 17-30 Å were determined by saturation recovery measurements of the enhancement of the nitroxyl spin lattice relaxation rates between ∼30-140 K, and by spin-echo measurements of the enhancement of spin-spin relaxation rates at 10-30 K. Interspin distances were also calculated, using the molecular modeling program Insight II (Accelrys, San Diego, CA). For most variants, distances determined from the temperature dependence of spin-echo intensities at a pulse spacing of 200 ns agree with distances measured by saturation recovery and calculated with Insight II within about an angstrom, which is within experimental uncertainties. Measurements of interspin distances via spin-spin relaxation enhancement have the advantages that maximum effects are observed for slower metal relaxation rates than are required for spin-lattice relaxation enhancement, and the impact diminishes as r⁻³ instead of r⁻⁶, as with spin-lattice relaxation enhancement, which permits measurements at longer distances.     

Mode‐coupling Smoluchowski dynamics of a double‐stranded DNA oligomer

The local dynamics of a double‐stranded DNA d(TpCpGpCpG)₂ is obtained to second order in the mode‐coupling expansion of the Smoluchowski diffusion theory. The time correlation functions of bond variables are derived and the ¹³C‐nmr spin–lattice relaxation times T₁ of different ¹³C along the chains are calculated and compared to experimental data from the literature at three frequencies. The DNA is considered as a fluctuating three‐dimensional structure undergoing rotational diffusion. The fluctuations are evaluated using molecular dynamics simulations, with the ensemble averages approximated by time averages along a trajectory of length 1 ns. Any technique for sampling the configurational space can be used as an alternative. For a fluctuating three‐dimensional (3D) structure using the three first‐order vector modes of lower rates, higher order basis sets of second‐rank tensor are built to give the required mode coupling dynamics. Second‐ and even first‐order theories are found to be in close agreement with the experimental results, especially at high frequency, where the differences in T₁ for ¹³C in the base pairs, sugar, and backbone are well described. These atomistic calculations are of general application for studying, on a molecular basis, the local dynamics of fluctuating 3D structures such as double‐helix DNA fragments, proteins, and protein–DNA complexes. © 1999 John Wiley & Sons, Inc. Biopoly 50: 613–629, 1999     

Accelerating proton spin diffusion in perdeuterated proteins at 100 kHz MAS

Fast magic-angle spinning (>60 kHz) has many advantages but makes spin-diffusion-type proton–proton long-range polarization transfer inefficient and highly dependent on chemical-shift offset. Using 100%-HN-[²H,¹³C,¹⁵N]-ubiquitin as a model substance, we quantify the influence of the chemical-shift difference on the spin diffusion between proton spins and compare two experiments which lead to an improved chemical-shift compensation of the transfer: rotating-frame spin diffusion and a new experiment, reverse amplitude-modulated MIRROR. Both approaches enable broadband spin diffusion, but the application of the first variant is limited due to fast spin relaxation in the rotating frame. The reverse MIRROR experiment, in contrast, is a promising candidate for the determination of structurally relevant distance restraints. The applied tailored rf-irradiation schemes allow full control over the range of recoupled chemical shifts and efficiently drive spin diffusion. Here, the relevant relaxation time is the larger longitudinal relaxation time, which leads to a higher signal-to-noise ratio in the spectra.     

Proton and phosphorus-31 magnetic relaxation studies on the interaction of polyriboadenylic acid with Mn2+

Proton and phosphorus-31 nuclear spin–lattice relaxation times T₁ and spin–spin relaxation times T₂ have been measured on the single-stranded polyriboadenylic acid [poly(A)]–Mn²⁺ system in a neutral D₂O solution in the temperature range 10°–90°C at 100 and 40.5 MHz, respectively, with the Fourier transform nmr method. Minimum values of T₁ have been found for all these nuclei, which have enabled the exact estimation of apparent distances from Mn²⁺ to H₂, H₈, H_1′, and the phosphorus nucleus to be 4.7, 4.1, 5.2, and 3.0 Å, respectively. The electron spin of Mn²⁺ penetrates into the phosphorus nucleus, giving ³¹P hyperfine coupling of more than 10⁶ Hz. Evidence of penetration of the electron spin into H₈ and H₂ is also obtained, suggesting direct coordination of nitrogen atoms of the adenine ring to the Mn²⁺ Ion. Combined with the result from proton relaxation enhancement of water, it is concluded that every Mn²⁺ ion added is bound directly to two phosphate groups with a Mn²⁺–phosphorus distance of 3.3 Å, while a part of the Mn²⁺ ions are simultaneouly bound to the adenine ring. It is estimated that 39 ± 13% and 13 ± 5% of Mn²⁺ are coordinated by N₇ and N₃ (or N₁), respectively. The motional freedom of poly(A) in the environment of the Mn²⁺ binding site has been found to be quenched to the extent that the rotational motion becomes several times slower than that of the corresponding Mn²⁺–free poly(A). The activation energies for the molecular motion are, however, practically unchanged from those for Mn²⁺–free poly(A), and are found to be 8.3, 8.5, 6.1, and 8.7 kcal/mol for H₈, H₂, H_1′, and phosphorus, respectively. T₂ of phosphorus is determined by the dissociation rate (k_?1) of Mn²⁺ from the phosphate group for the whole temperature range studied with activation enthalpy of 6.5 kcal/mol. The dissociation rates of Mn²⁺ from the adenine ring are also estimated from proton T₂ values below 50°C.     

Hemoglobin aggregation in oxygenated sickle cells studies by carbon-13 rotating frame spin-lattice relaxation in the presence of an off-resonance radiofrequency field

Evidence is presented which shows that hemoglobin S in sickle cells has a tendency to aggregate even in the oxygenated state. The basis for that conclusion is derived from ¹³C-nmr rotating-frame spin–lattice relaxation studies in the presence of an off-resonance radiofrequency field in which the carbonyl resonances of hemoglobins in erythrocytes are examined. The experiments indicate that the rotational correlation time of hemoglobin S in oxygenated sickle cells at 38°C is 130 nsec compared to a value of 95 nsec for hemoglobin A in normal erythrocytes at the same temperature and the same mean cell hemoglobin content.     

Cross-correlated relaxation rates between protein backbone H–X dipolar interactions

The relaxation interference between dipole–dipole interactions of two separate spin pairs carries structural and dynamics information. In particular, when compared to individual dynamic behavior of those spin pairs, such cross-correlated relaxation (CCR) rates report on the correlation between the spin pairs. We have recently mapped out correlated motion along the backbone of the protein GB3, using CCR rates among and between consecutive H^N–N and H^α–C^α dipole–dipole interactions. Here, we provide a detailed account of the measurement of the four types of CCR rates. All rates were obtained from at least two different pulse sequences, of which the yet unpublished ones are presented. Detailed comparisons between the different methods and corrections for unwanted pathways demonstrate that the averaged CCR rates are highly accurate and precise with errors of 1.5–3% of the entire value ranges.     

Structural and dynamics characterization of norovirus protease

Norovirus protease is an essential enzyme for proteolytic maturation of norovirus nonstructural proteins and has been implicated as a potential target for antiviral drug development. Although X‐ray structural studies of the protease give us wealth of structural information including interactions of the protease with its substrate and dimeric overall structure, the role of protein dynamics in the substrate recognition and the biological relevance of the protease dimer remain unclear. Here we determined the solution NMR structure of the 3C‐like protease from Norwalk virus (NV 3CLpro), a prototype strain of norovirus, and analyzed its backbone dynamics and hydrodynamic behavior in solution. ¹⁵N spin relaxation and analytical ultracentrifugation analyses demonstrate that NV 3CLpro is predominantly a monomer in solution. Solution structure of NV 3CLpro shows significant structural variation in C‐terminal domain compared with crystal structures and among lower energy structure ensembles. Also, ¹⁵N spin relaxation and Carr–Purcell–Meiboom–Gill (CPMG)‐based relaxation dispersion analyses reveal the dynamic properties of residues in the C‐terminal domain over a wide range of timescales. In particular, the long loop spanning residues T123–G133 show fast motion (ps‐ns), and the residues in the bII–cII region forming the large hydrophobic pocket (S2 site) undergo conformational exchanges on slower timescales (μs–ms), suggesting their important role in substrate recognition.     

Determination of NOE "silent" dipolar interactions between magnetically equivalent nuclei: application to poly(dA).poly(dT)

The ratio of the spin–spin relaxation rate, R₂, to the selective spin–lattice relaxation rate, R, can be used to detect and determine the magnitude of dipolar interactions between magnetically equivalent unclei in macromolecules. We use this method to show that there is strong interaction between adjacent AH2 protons in poly(dA) · poly(dT). The largest possible AH2–AH2 distance that will account for the observed magnitude of the interaction is ～3.5 Å, provided the AH2 protons are located close (within 0.6 Å) to the helix axis. Structures with shorter AH2–AH2 separations could account for the nmr data, but they are rejected because they would reduce the interbase separation to unacceptably small values. A comparison of rates measured at 360 and 500 MHz shows that the chemical-shift anisotropy makes a negligible contribution to the observed relaxation rates.     

31P Magnetic relaxation studies of yeast transfer RNAPhe.

The molecular mechanism of thermal unfolding of yeast tRNA^Phe in 20 mM NaCl, 1 mM EDTA, and 10 mM MgSO₄, pH 7.1 ± 0.1, has been examined by ³¹P magnetic relaxation and the nuclear Overhauser effect methods at 40.48 MHz in the temperature range of 22.5–80°C. Two partially resolved ³¹P resonance peaks of yeast tRNA^Phe have been found to behave distinctively different in their longitudinal relaxation times. Individual intensities of the two partially resolved peaks have been quantitatively estimated by the use of relaxation data and the nuclear Overhauser effect as a function of temperature. The results of these observations largely support the earlier suggestion by Guéron and Shulman that the high- and low-field parts of the main ³¹P resonance cluster originate from phosphorus nuclei belonging to the double-helical and nonhelical regions of the tRNA, respectively. The spin-lattice relaxation of the phosphorus nucleus has been found to be determined dominantly by the dipolar interaction with the surrounding ribose protons at this observing frequency. Rotational correlation times for the two portions of the ribose-phosphate backbone of the tRNA have been separately deduced from the quantitative treatment of the ³¹P nuclear spin-lattice relaxation times (T₁) and the nuclear Overhauser effect. The result indicates that the two portions undergo internal motions at distinctively different rates of 10⁸–10¹⁰ sec^?1 order in the temperature range of 22.5–80°C, and that the thermal activation of these motions occurs at least in three distinctive steps, i.e., 22.5–31, 31–40, and 40–80°C. The rates of the internal motions and the associated activation energies in respective steps give some insight into the thermo-induced change of the yeast tRNA^Phe structure.