Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Jan 14.
Published in final edited form as: J Am Chem Soc. 2009 Jan 14;131(1):118–128. doi: 10.1021/ja8045926

Observation of a Low-Temperature, Dynamically Driven, Structural Transition in a Polypeptide by Solid State NMR Spectroscopy

Vikram S Bajaj 1,a, Patrick CA van der Wel 1,b, Robert G Griffin 1,*
PMCID: PMC2651395  NIHMSID: NIHMS91489  PMID: 19067520

Abstract

At reduced temperatures, proteins and other biomolecules are generally found to exhibit dynamic as well as structural transitions. This includes a so-called protein glass transition that is universally observed in systems cooled between 200–230K, and which is generally attributed to interactions between hydrating solvent molecules and protein side chains. However, there is also experimental and theoretical evidence for a low-temperature transition in the intrinsic dynamics of the protein itself, absent any solvent. Here, we use low-temperature solid state NMR to examine site specific fluctuations in atomic structure and dynamics in the absence of solvents. In particular, we employ magic angle spinning NMR to examine a structural phase transition associated with dynamic processes in a solvent-free polypeptide, N-f-MLF-OH, lattice at temperatures as low as 90K. This transition is characterized by the appearance of an extra set of lines in 1D 15N spectra as well as additional cross peaks in 2D 13C-13C and 13C-15N spectra. Interestingly, the gradual, temperature-dependent appearance of the new spectral component is not accompanied by the line broadening typical of dynamic transitions. A direct comparison between the spectra of N-f-MLF-OH and the analog N-f-MLF-OMe, which does not display this transition, indicates a correlation of the structural transition to the temperature dependent motion of the aromatic phenylalanine side chain. Several quantitative solid state NMR experiments were employed to provide site-specific measurements of structural and motional features of the observed transition.

Introduction

The role of dynamics and flexibility in protein function has been clear since the very first crystal structure1. In order to bind substrate or channel a chemical reaction forward, an enzyme assumes one or more conformational states in sequence, such that energy is transduced in the desired direction. Transitions between these local energy minima, which are near the global minimum of the protein energy landscape, are mediated by concerted dynamical rearrangements at ambient temperatures 2–4. At the same time, a protein must be sufficiently robust and stable to reliably adopt a unique conformation or set of conformations that are essential for its function. This balance between conformational flexibility and structural robustness has evolved in each protein to allow it to function in its native environment. This has been experimentally observed as a temperature-dependent loss or gain of activity for various biological systems that correlates to specific changes in protein flexibility5–7. This is true not only for globular proteins, but also for membrane protein-lipid complexes such as photosystem II8 and bacteriorhodopsin9. Interestingly, the temperature at which this loss of dynamics and function occurs appears largely independent of the type of system. This seemingly universal transition occurs at 200 – 230 K, and is known as the ‘protein dynamical transition’ or the ‘protein glass transition’ 10–15. The latter name relates to the observed analogy with the dynamics encountered in disordered media such as liquids. A similar transition has even been observed in other biological macromolecules 16.

There have been a number of studies that have attempted to characterize the physical basis of this process. Experimental data from x-ray crystallography 17, 18 and FTIR 19, and molecular dynamics simulations 12 support the idea that the protein glass transition is due to a coupling of the protein motions with the dynamics of the solvent. The dynamic modes of the bulk water molecules, in particular, appear to influence the motion of the polypeptide chain; ordering of water at low temperatures is then associated with a loss of this dynamically driven activation 17 through a solvent caging effect. At temperatures slightly above the glass transition, ordering of bound waters results in increased local viscosity and prevents the anharmonic and concerted protein motions that give rise to mechanistically significant rearrangements. This has been supported by studies showing that the transition temperature can be modulated by changes in the viscosity of the solvent system 20–22.

Aside from these solvent-dependent contributions to the protein glass transition13, 16 22–24 more recent data have highlighted residual anharmonic motions occurring below 200K, and showed that these motions appear independent of the hydration state. This solvent-independence has been observed in neutron scattering experiments on dehydrated myoglobin 25, lysozyme 16, 26 and photosystem II 24, molecular dynamics simulations in the absence of solvent 27, and the extrapolated results of solution state NMR experiments at ambient temperatures 13. Recent molecular dynamics simulations by Vitkup and co-workers 12 suggest that solvent mobility dominates the glass transition down to 180 K, and that intrinsic protein dynamics play the dominant role at lower temperatures. Therefore, the onset of anharmonic motions near 100K appears to be an intrinsic property of the protein conformational energy surface alone. It was suggested that these dynamics are dominated by the onset of thermally activated methyl group rotations 13, 16, 23. Interestingly, previous solid state 2H NMR studies of −CH3 group motion in Val and Ala have shown that thermally activated threefold hops initiate at 110K and 133K respectively 28, 29.

Clearly, the low temperature dynamic behavior of proteins is more complex than was initially assumed and a detailed understanding is important for several reasons. First, it will provide insight into the concerted dynamic rearrangements that are necessary for optimal biological function. Second, from a practical perspective, a number of biophysical techniques, including x-ray crystallography, solid state NMR (SSNMR) and cryo-electron microscopy, frequently require cooling of the sample to temperatures below the protein glass transition temperature. Since the protein in this state is not biologically active, it is important to understand how measured structural parameters change as a function of temperature. Finally, the low temperature dynamics, if better understood, may allow for trapping of mechanistically relevant conformational intermediates 7, 11.

Here, we demonstrate the utility of high resolution, temperature dependent magic angle spinning (MAS) NMR spectroscopy in a study of the intrinsic dynamics and structure of the chemotactic tripeptide, N-f-MLF-OH in the absence of solvent. These tripeptide crystals form a model system for SSNMR experiments and have been well-characterized in previous experiments 32–35, resulting in an early solid-state NMR based structure 36. In the current context, the system is particularly appropriate because it crystallizes in solvent-free microcrystals and yet experiences a representative range of dynamic phenomena. Indeed, a variety of conformational dynamics are known to be present in the lattice, including rotamer interconversion, librational motion of side chains, phenylalanine ring flips, and three-fold “hopping” of methyl groups, representing a variety of amino acid dynamics observed in proteins in general. Further, the addition of a single methyl group produces N-f-MLF-OMe, which adopts a similar crystal structure (Figure 1), as reflected by its similar chemical shifts and powder diffraction patterns. In addition, and in contrast to N-f-MLF-OH, the crystal structure of N-f-MLF-OMe was solved by x-ray crystallography 30. However, the latter peptide has one salient structural difference in that its Phe aromatic ring is sterically restricted and thus hindered from flipping. Therefore, in the 13C MAS spectrum of N-f-MLF-OMe recorded at 300 K, we observe all six lines from the ring, whereas in N-f-MLF-OH we see lines only from the Cγand Cζ 13C’s. The lines from Cεand Cδare attenuated due to the flipping motion interfering with the 1H-13C decoupling37,38.

Figure 1.

Figure 1

Open in a new tab

Structure of N-f-MLF-OMe crystal lattice30. Panel (a) shows a ‘top-down’ view of the crystal lattice. Panel (b) shows the steric crowding and relative orientation of the phenylalanine side chains, allowing π−π aromatic interactions. This figure was prepared using PyMOL31.

Solid state NMR is a robust method of ab initio structure determination that can concurrently provide detailed information about motional processes over a wide range of time scales. Furthermore, there has been significant progress in the development of low-temperature MAS instrumentation in recent years, facilitating an investigation of the protein glass transition 39, 40. In 1D spectra collected as a function of temperature, the effects of dynamics are readily observed. They include classical conformational exchange and interference phenomena that are manifest when dynamic processes occur at rates similar to the magic angle spinning or 1H decoupling. The tripeptide N-f-MLF-OMe provides a rigid reference structure, showing little effect of cooling to temperatures as low as 90K. In contrast, similar cooling causes dramatic changes in the NMR spectra of the N-f-MLF-OH crystals. These changes include the emergence of a structurally distinct substate of the peptide near the glass transition temperature, which we show is correlated to an unusual dynamically driven structural transition occurring between ~90 and ~200 K. The distinct structural forms of the peptide were also characterized using solid state NMR methods previously applied to N-f-MLF-OH at room temperature36 and, more recently, in amyloid fibrils from transthyretin 41. This strategy involves the application of dipolar recoupling experiments in uniformly labeled systems to measure heteronuclear 13C-15N distances 34, 42, 43 and tensor correlation experiments 33,44–52 to constrain backbone torsion angles. Based on our experimental data we report on the change of dynamics and the correlated changes in structure. The observed dynamic/structural transition is also discussed in terms of its possible relation to the physical origins of intrinsic low temperature protein motions near the protein glass transition.

Methods

Sample preparation

Unlabeled tripeptides N-formyl-Met-Leu-Phe-OH (N-f-MLF-OH) and N-formyl-Met-Leu-Phe-OCH3 (N-f-MLF-OMe) were obtained from Bachem (King of Prussia, PA). Selectively deuterated N-f-MLF-OH containing ring-deuterated Phe-d5 was prepared by solid phase peptide synthesis by SynPep Corporation (Dublin, CA). Uniformly labeled N-f-[U-13C,15N-MLF]-OH was synthesized by solid phase peptide synthesis by CS Bio Inc. (Menlo Park, CA). Isotopically labeled Phe-d5, FMOC-[U-13C,15N-Met], FMOC-[U-13C,15N-Leu], and FMOC-[U-13C,15N-Phe] were obtained from Cambridge Isotope Laboratories (Andover, MA). N-f-MLF-OH and N-f-MLF-OMe were crystallized from isopropanol and benzene, respectively, in accordance with previously published protocols 30, 53. Small, needlelike crystals were obtained after dissolution in warm solvent followed by slow evaporation to dryness. Distance and torsion angle measurements were conducted in a sample diluted to 10% in natural abundance peptide prior to crystallization in order to minimize the effects of intermolecular dipolar couplings.

NMR measurements and analysis

Magic angle spinning experiments were performed using a home-built spectrometer operating at 380 MHz 1H frequency (designed by D. J. Ruben, Francis Bitter Magnet Laboratory, MIT) and a triple channel (1H, 13C, 15N) 4 mm MAS probe designed for low temperature dynamic nuclear polarization (DNP) experiments39. Sample cooling was performed by using pre-cooled nitrogen gas to provide the drive and bearing pressure for MAS. The spinning frequency of 6.25–7 kHz was controlled by a Bruker MAS controller (Billerica, MA). At each temperature, 13C chemical shifts were referenced relative to dilute, aqueous DSS using external referencing via the 13C chemical shifts of adamantane 54, and 15N chemical shifts were referenced indirectly to liquid ammonia using the suggested IUPAC frequency ratios (13C/1H) of aqueous DSS and liquid NH3 (15N/1H) 55, 56. Spectra were processed and visualized using the NMRPipe 57, Sparky58, and CCPNMR 59–62 software packages.

Static deuterium NMR spectra were obtained on a home-built spectrometer operating at 61 MHz 2H frequency, using a quadrupolar echo pulse sequence63 with 2 µs π/2 pulses, a 35 µs echo delay and a 8–12 s recycle delay. The experimental lineshapes were simulated with an updated version of the 2H line shape package TURBOPOWDER64 in order to determine the mechanism and rate of the Phe ring motion. The spectra from the δ and ε 2H’s were simulated separately from those of the terminal ζ deuteron and the resulting spectra combined, allowing lineshape and intensity to be matched to the experimental variable temperature data.

Structural measurements

Details concerning the MAS SSNMR pulse sequences used for structural measurements are included in the supporting information. 15N-13C distances were measured using the 3D TEDOR pulse sequence43. The data were fit using an approximate analytical model of the spin dynamics for multiple, coupled spins under TEDOR 43 using a Levenberg-Marquardt algorithm for minimization of the χ2fit parameter.

Backbone torsion angle measurements were performed using HNCH and HNi+1CiH tensor correlation experiments44,52. The correlated dephasing under 1H-15N and 1H-13C couplings is monitored through the intensity of the 15N-13C cross peaks as a function of the synchronously incremented dipolar mixing time. The relative rate of incrementation was chosen to optimize the sensitivity of the experiment (in this case, n=2). In these experiments, TMREV-4 was applied at a spinning frequency of 8.333 kHz65. The 3D HCCH experiment66 for side chain torsion angle measurements was implemented as shown in the supporting information. Double quantum coherence, generated with band selective homonuclear recoupling sequence SPC53, dephases under the recoupled 1H-13C dipolar interactions and can be used to constrain the χ angle67. In the related NCCN experiment46,51,68, which provides constraints on the ψ backbone torsion angle, SPC53 was used to generate double quantum coherence between C’-Cα. Here, REDOR42 is used to simultaneously recouple the N-C’ and N-Cα interactions, causing ψ-dependent dephasing of the DQ coherence. In a future publication we plan to describe a detailed structural study of both forms of the N-f-MLF-OH peptide described here.

Results

Room temperature measurements

Figure 2 shows the 1D solid-state MAS NMR spectra at room temperature for both N-f-MLF-OH and N-f-MLF-OMe. In N-f-MLF-OMe the chemical shifts of each 13C in the phenyl ring of N-f-MLF-OMe are distinct, indicating a static structure without dynamics on the NMR time scale. The ring in the N-f-MLF-OH peptide, on the other hand, is undergoing 180-degree flips, averaging the chemical shifts of Cδ and Cε ring 13C’s. Due to interference of the ring dynamics with the heteronuclear 1H-13C decoupling37, 38, these carbon signals are broadened, unless the Phe side chain is deuterated (compare panels (a) and (b)), thus eliminating the strong heteronuclear and 1H-1H dipolar couplings.

Figure 2.

Figure 2

Open in a new tab

Assignment of one-dimensional 13C and 15N MAS spectra of N-f-MLF-OH and N-f-MLF-OMe recorded at room temperature: (a) 13C spectra of natural abundance N-f-MLF-OH; (b) unlabeled N-f-MLF-OH with a 2H-labeled side chain (Phe-d5); (c) natural abndance N-f-MLF-OMe; (d) 15N spectra of N-f-[U-13C,15N]MLF-OH; (e) natural abundance 15N spectrum of N-f-MLF-OMe. Assignments for N-f-MLF-OMe were made in analogy with N-f-MLF-OH, with ambiguities denoted with italicized labels.

Temperature dependence

Both peptides were examined by 13C and 15N magic angle spinning NMR over the temperature range 90–300 K and the spectra revealed remarkable differences in behavior. This is most obvious in the 1D 15N spectra illustrated in Figure 3, where on cooling natural abundance N-f-MLF-OMe from 300 K to 90 K the spectra do not change significantly. Since the 15N chemical shift is a sensitive indicator of structure and dynamics, it appears that there is little change in either property. In contrast, the 15N spectra of the (isotopically labeled) N-f-MLF-OH analog show a significantly different pattern (Figure 3b). Upon cooling below 200K, a second set of lines appears and co-exists with the original peaks (observed at room temperature), but diminishes again in intensity as the sample is further cooled to 90K or below. As illustrated in Figure 3c, spectra acquired when the sample is warmed from 90 K to 243 K reveal a significant hysteresis in the relative intensities of the lines from the two coexisting forms.

Figure 3.

Figure 3

Open in a new tab

Variable temperature 15N spectra of (a) natural abundance N-f-MLF-OMe and (b,c) 10% [U-13C, 15N] N-f-MLF-OH. (a) No spectral changes are observed when cooling unlabeled N-f-MLF-OMe. (b) Cooling N-f-MLF-OH below 200 K yields spectra with extra lines. (c) Spectra obtained upon warming N-f-MLF-OH from 90K to room temperature illustrating the hysteresis in the phase transition.

The temperature-dependent behavior was also examined using 13C MAS NMR. As shown in Figure 4, there are no changes in the 13C chemical shifts for the N-f-MLF-OMe 13C sites as a function of temperature. However, we do observe significant changes in line widths and intensities for several lines. The most dramatic perturbations are to the resonances corresponding to the two methyl groups of the Leu side chain (indicated with arrows in the Figure 4d). The observed intensity loss is due to the fact the rate of threefold motion of the −CH3’s is comparable to the decoupling field 37,38. Note that other methyl groups in the Met side chain and the -OMe at the C-terminus (marked with * in Figure 4) experience less steric hindrance and thus do not exhibit these intensity losses.

Figure 4.

Figure 4

Open in a new tab

Variable temperature 1D spectra of unlabeled N-f-MLF-OMe. (a–d) 13C and (e–h) 15N spectra recorded at 298, 190, 155, and 95 K. Data were acquired at ωr/2π = 6.25 kHz. Arrows indicate Leu methyl resonances that are most strongly attenuated at low temperature. Less restricted methyl resonances, indicated with ‘*’, remain unaffected.

In the case of the Leu and Met methyl groups of N-f-MLF-OH, we observed similar behavior, with significant intensity losses in the Leu methyl resonances, and less effect on the Met methyl resonance. However, as illustrated in Figure 5 with 13C spectra of the natural abundance peptide, we now also observe substantial changes in the number of observed signals and their apparent line widths when cooling below 200K, coinciding with the appearance of the additional signals in the 15N spectra the N-f-MLF-OH (Figure 3). In particular, as shown in Figure 5b, we observe the formation of an additional, distinct spectroscopic ‘form’ of the peptide between 200K and 90K. Note also that at 200K and below, the resonances of the Phe aromatic ring increase in intensity, and at 90 K additional individual peaks are visible indicating that the rate of 180° ring flipping has ceased. Finally we observe that the resolution in the spectrum is largely restored at 90 K, as illustrated in Figure 5a(bottom), and that most resonances exhibit 13C chemical shifts similar to the values at room temperature.

Figure 5.

Figure 5

Open in a new tab

Variable temperature 1D spectra of unlabeled N-f-MLF-OH. Column (a) shows the coarse temperature dependence between 290 K and 90K. (b) Spectra illustrating the spectral changes and apparent linebroadening associated with the transition between 200K and 110K.

Assignment experiments

We performed two-dimensional chemical shift correlation experiments to generate site-specific assignments at 298K, 175K, and 90K. The results of these measurements are summarized in Figure 6–Figure 8.Figure 6 shows the 2D 13C-13C correlation spectra recorded at each temperature. These spectra alone provide complete 13C assignments of all resonances, with the exception of those that disappear upon cooling (e.g. various methyl signals). We observe a single form of the peptide at 298K and 90K, but a mixture of two forms at 175K. Of these, one ‘persistent’ form has chemical shifts which resemble those of the room temperature and 90K forms of N-f-MLF-OH, while the other form is a novel, ‘transitional,’ form that arises as the sample approaches a temperature of 175K. Figure 7 highlights the aliphatic Cα-Cβ region, where the correspondence in chemical shift is most apparent. It is easy to see from these spectra that the cross peak widths in the 175 K spectrum are essentially identical to those observed at 298 K and 90 K. Therefore, we conclude that the apparent loss of resolution in the 1D spectra, for instance in the Cα region near 50–60 ppm, is caused by the emergence of this second structural form at intermediate temperatures rather than by line broadening due to conformational disorder. Heteronuclear correlation spectroscopy measurements served to provide sequential assignments, thereby assigning the 15N chemical shifts and confirming our 13C assignments. The NCACX and NCO spectra at 175K are shown in Figure 8, along with their cross peak assignments. Based on these assignments, we could assign the extensive variable temperature 1D data (see e.g. Figure 2–Figure 4), track the change in chemical shift as a function of temperature, and monitor the appearance of the transitional form below 200K. Tables containing the chemical shifts and assignments are available as supporting information.

Figure 6.

Figure 6

Open in a new tab

13C-13C correlation spectra of 10% U-13C,15N labeled N-f-MLF-OH at (a) 298K, (b) 175K, and (c) 90K. A DARR69–71 mixing time of 10ms resulted in single bond, and weaker two-bond correlations. The top and bottom panels show aliphatic-aliphatic and carbonyl-aliphatic cross-peaks, respectively. For clarity, cross-peaks from the novel, ‘transitional’ form seen at 175K are highlighted in red.

Figure 8.

Figure 8

Open in a new tab

2D 15N-13C correlation spectra on [10% U-13C,15N N-f-MLF-OH] at 175K and 8 kHz MAS: NCACX (a), with 10ms DARR 13C-13C mixing, showing intra-residue assignments (asterisks mark several weak inter-residue correlations); NCO 2D (b) showing sequential N-C’ cross-peak assignments.

Figure 7.

Figure 7

Open in a new tab

Expanded regions from the experimental spectra shown in Figure 6, highlighting the 13Cα-13C β cross peaks, at (a) 298K, (b) 175K, and (c) 90K.

Characterization of the dynamics

The difference in behavior between the two peptides, namely the presence or absence of a transitional form, appears to correlate with differences in their dynamics. For both peptides, we observe the expected three-fold hopping motion of the methyl groups, with a loss of intensity in the Leu methyl groups at reduced temperatures. Further, in N-f-MLF-OH we observed evidence of Phe ring motion, which is absent in N-f-MLF-OMe. Therefore, we also examined the molecular dynamics through a number of solid state MAS experiments. The 1H-13C dipolar interaction is attenuated by any dynamics and was therefore measured using the TMREV recoupling sequence65. The 1H-13C dipolar couplings of several 13C-1H groups change significantly as a function of temperature (see supporting information). At room temperature the methyl dipolar couplings are averaged due to a three-fold hopping motion, and, as the temperature is reduced, the methyl motions slow and their C-H dipolar couplings increase until they reach a limit near 200–225K. The temperature-dependence for the magnitude of the scaled dipolar coupling is also found to be residue dependent, consistent with the observations in the 1D 13C spectra. There is also a large change in the behavior of the aromatic carbon sites, which are mobile down to 250K but progressively slow down as the temperature approaches 165K.

The motion of the phenyl ring was examined in more detail with 2H NMR quadrupole echo experiments on crystals of N-f-MLF-OH containing a deuterated phenyl ring (Phe-d5).Figure 9(a) shows the resulting 2H 1D powder spectra at a number of temperatures. Fits of the powder lineshape (panel (b)) and of the quadrupolar echo intensity data (panel (c,d)) indicate that the ring is executing 180° flips at a rate that varies from <102 s−1 near 200K to ~105 s−1 near room temperature. The latter was confirmed with room temperature inversion recovery measurements (shown in panel (e)). An Arrhenius analysis of the estimated hopping rates reveals an activation energy for ring flip of ~38 kJ/mol, which is similar to the activation energies observed for a number of other crystalline Tyr- or Phe-containing peptides 72,73.

Figure 9.

Figure 9

Open in a new tab

Static deuterium NMR data for N-f-ML[F-d5]-OH. (a) 1D spectra (left) at 198 to 316K together with simulated spectra (b) for estimated Phe flipping rates, (c) overall signal intensity of observed signal at 198 to 316 K, (d) simulated flipping rate-dependence of signal intensity for Phe-d5, with the dashed box indicating the range of conditions matching the experimental results, and (e) room temperature inversion recovery measurement with simulated curve at 3.5 × 105 Hz flipping rate.

Structural measurements

In order to examine the structural features of the two forms of N-f-MLF-OH that co-exist at 175K, we used dipolar recoupling experiments to measure heteronuclear 13C-15N distances 34,42,43 and tensor correlation experiments 33,44–52 to constrain backbone and side chain torsion angles. All measurements were conducted using a sample consisting of uniformly 13C,15N-labeled N-f-MLF-OH diluted to 10% in natural abundance peptide prior to crystallization, to minimize the effects of intermolecular dipolar couplings. Distance constraints were obtained via 3D TEDOR43 experiments performed at 298, 175 and 90K. This resulted in numerous intramolecular 13C-15N distances, which are tabulated in the supporting information. An example of the fit for weak dipolar couplings corresponding to distances of 5.4 Å and 6.4 Å (Met-N-Cε) is shown in Figure 10(a). Note that the approximate analytical model used to fit these data neglects the orientation dependence of the spin dynamics and hence introduces a systematic error of approximately 10%–20%, which is greater than the random error in these data.

Figure 10.

Figure 10

Open in a new tab

(a) Example of fitting (solid lines) of long distances in TEDOR experiments, showing the Met N-Cε distance at 175K. Experimental data are shown as open circles (persistent form) and solid triangles (transitional form). (b) Comparison of 175K and 298K N-C distances. The transitional form (MLF)’ has systematically longer distances than the persistent form (MLF), which closely resembles the room temperature structure. The solid and dashed lines indicate the linear correspondence for the room temperature and ‘persistent’ form at 175K.

These 15N-13C distances allow initial structural comparisons of the co-existing 175K forms, and provide insights into the localization of the largest changes. Figure 10(b) shows a comparison of both forms to the room temperature structure, indicating that the persistent form is structurally similar to the room temperature form. The transitional structure is characterized by an overall increase in the measured 13C-15N distances (see also Figure S2 in the supporting information). While the distances in the persistent structure vary at most 8%(or 0.4 Å) from the equivalent distance at 298K, the transitional structure displays much larger changes of as much as 45%(or 1.3 Å).

To obtain further structural information, side chain and backbone torsion angle were measured using a number of techniques. These experiments incorporate the TMREV sequence for γ-encoded heteronuclear recoupling of the 1H-13C dipolar coupling65 or REDOR for recoupling of the 15N-13C dipolar interaction42. A 3D HCCH experiment allowed the simultaneous examination of multiple side chain torsion angles in the co-existing N-f-MLF-OH forms found at 175K. The measurement resulted in dephasing trajectories (Figure 11) suggesting very similar Leu side chain conformations, but a significant conformational difference in the Met side chain. The backbone ϕ and ψ torsion angles were accurately constrained through HNCH and HNi+1CiH experiments, with secondary constraints on ψ through NCCN dipolar correlation experiments. Representative data for the Leu residue are shown in Figure 12 (additional data are in the supporting information). The figure also includes simulated fit curves obtained in the analysis of these torsion angle data. These fit results illustrate the complementary angular sensitivity for the various backbone torsion angle techniques. Solutions for the torsion angle are inherently dependent on the precise value of the heteronuclear 1H-15N and 1H-13C couplings and on the three-atom bond angles used in the calculation of the projection angles. For all fits, we used the equilibrium parameters specified in the X-Plor NIH74 NMR refinement force-field for consistency. The expected variation of bond angles about these average values can induce a ±10–15° uncertainty in the solution for the projection angle, which in all cases exceeds the random error in these experiments. The calculated torsion angles are included as supporting information, and indicate that the backbone conformation is similar between the two structural forms.

Figure 11.

Figure 11

Open in a new tab

Experimental data from HCCH tensor correlation experiment designed to measure N-f-MLF-OH side chain torsion angles at 175K: (a) Leu χ1, (b) Met χ1, and (c) Met χ2. The results show that the Met side chain conformation is different between the two forms of MLF which co-exist at 175K.

Figure 12.

Figure 12

Open in a new tab

Examples of data fitting and resulting solution surfaces for the backbone torsion angles of Leu in the both forms at 175K. Measurements involved ϕ via measured via HNCH (a,d), and ψ via HNi+1CiH (b,e), and NCCN (c,f) tensor correlation experiments.

Discussion

We have obtained detailed experimental data on the dynamic and structural behavior of two microcrystalline tripeptides as a function of temperature. These polypeptide lattices are devoid of water or organic solvent molecules (tested explicitly via Karl Fischer water analysis at Galbraith Laboratories, Inc., Knoxville, TN; results not shown) and therefore all observations reflect intrinsic polypeptide dynamics. Consistent with the absence of solvent and established features of dehydrated protein dynamics, there was no evidence of a dynamic or structural transition between 200 and 240K. As discussed in the introduction, this is the typical regime where the solvent-dependent protein glass transition takes place. However, our NMR data do show a structural transition specific to one of the peptides at temperatures between 90 and 200K. This appears to reflect a distinct difference in motional behavior between the two tripeptide crystal lattices, despite their similar chemical structure and crystallization behavior. Specifically, the O-methylated peptide N-f-MLF-OMe forms a largely rigid system and temperature-dependent changes in its NMR spectra are relatively small and confined to threefold hopping of the methyl groups which is manifest in large changes in their 13C signal intensities with temperature. We further noted significant variations in methyl group dynamics between different residues that are correlated to the extent of steric hindrance to the methyl group three-fold jump. The motional interference with the 1H-13C decoupling37,38 due to a reduction in CH3 jump rate occurs at a higher temperature for the Leu side chain methyls than for the Met CH3 group, which remains mobile down to 90K, preventing such destructive interference. The 15N and 13C resonance chemical shifts are basically temperature independent in N-f-MLF-OMe, including those of the Phe side chain (which is immobile even at room temperature).

Spectra of the N-f-MLF-OH peptide reveal similar methyl group dynamics (disappearance of Leu methyls). Additional measurements of the CH3 motions based on the reduction in the 13C-1H dipolar coupling reinforce the observation of residue-type dependence, with substantial modulation of the temperature-dependence due to differences in the steric environment. Unlike the O-methylated peptide, there are significant additional structural and dynamical changes upon cooling. These effects include the temperature-correlated emergence of a second conformational substate that arises below 200K, and then eventually disappears upon cooling to 90K. We applied several solid state NMR methods to measure structural parameters of the peptide at 175K, where this ‘transitional’ form co-exists with a ‘persistent’ form that is present at both room temperature and 90K.

The measured distance and backbone torsion angle data (tabulated in supporting information) were used for structure refinement, employing protocols for simulated annealing molecular dynamics41 as implemented in the program X-Plor-NIH74. Due to the absence of side chain constraints at 90K, no structural refinement was attempted for N-f-MLF-OH at this temperature. In addition, conformational ensembles obtained from refinement at 175K are presently of insufficiently high resolution to allow a detailed structural analysis. Thus, we are continuing these experiments and will report them in a future publication. Nevertheless, we can draw some conclusions about the structural changes during the dynamically driven transition. The backbone structure was found to be preserved (as indicated by the backbone torsion angle measurements). The experimental data indicate changes in the conformation of the side chains, in particular the Met side chain. This is apparent from the torsion angle measurements (Figure 11), where a significant change is observed in the Met χ2 angle. Similar conclusions can be drawn from the distance measurements, as illustrated in Figure 13. This figure shows the room temperature SSNMR structure of N-f-MLF-OH36 with lines indicating the measured internuclear distances. The distances that deviate from their room temperature values are highlighted in color, alongside the change in distance in angstroms. Also here, a large consistent change is occurring in the Met side chain. In addition, we observed a significant increase in the Leu N-Cγ/Cδ1 distance (due to overlap, the carbon sites could not be distinguished), suggesting that the Leu side chain is oriented differently in the transitional form. Since the Leu χ1 angle was found to be the same between the two forms (figure 11), the difference must be due to a change in the Leu χ2 angle (analogous to Met).

Figure 13.

Figure 13

Open in a new tab

Illustration of the deviation from the room temperature distances34, 36 for each of the measured 15N-13C measurements at 175K, (a) for the persistent form (MLF) and (b) the transitional form (MLF)’. Each of the observed distances is indicated by a line, with those that deviate by more than 0.25 Å highlighted and marked with the deviation (in Å). The largest deviations are located in the Met side chain. This figure was prepared using UCSF Chimera 75.

Let us now compare N-f-MLF-OH with N-f-MLF-OMe and their behavior upon cooling. The room temperature monomer structure of N-f-MLF-OH was previously determined by solid state NMR 36 and was found to resemble the X-ray structure of N-f-MLF-OMe 30. Further, X-ray powder diffraction data suggest that the unit cell and space group of both peptide crystals are similar (see supporting information). These structural similarities indicate that the spectral differences are due to differences in the dynamics and that the transition represents a low-temperature dynamic transition, correlated to a structural transition (i.e. one which does not occur in the absence of these dynamics). In the conventional picture of temperature-activated conformational exchange as monitored by NMR, a transition between two states results in spectra with an averaged, narrow resonance at high temperature, a broad, single resonance at intermediate temperatures, and separate narrow resonances at temperatures low enough to arrest the dynamics. Interestingly, N-f-MLF-OH does not display this behavior, but rather shows a distribution between two distinct, similarly narrow sets of NMR lines at intermediate temperatures. Even more atypical seems the observation that the form that appears upon cooling, subsequently disappears at even lower temperatures.

Indeed, the phase transition in N-f-MLF-OH takes place between 200 and 100K where, concomitantly, there is a drastic reduction in the intensity of the methyl group resonances. The observed dynamic event does correlate with the observed onset of anharmonic motion at temperatures above 100K, which was proposed to be associated with methyl group motion 12, 16, 26, 76. However, our system highlights the fact that another potentially important source of mobility in proteins is the aromatic side chains, such as phenylalanine. Specifically, the methyl dynamics are present in N-f-MLF-OMe as well and therefore do not account for the observed transition. In fact, the most visible difference between the N-f-MLF-OH and N-f-MLF-OMe crystals is in the rate of reorientation of the Phe ring. As illustrated in Figure 1, the aromatic ring in N-f-MLF-OMe is prevented from flipping by the close proximity of a number of bulky groups, including the C-terminal methyl group (OMe). In contrast, the 13C and 2H spectra clearly show that the Phe ring in N-f-MLF-OH is executing 180° flips. This suggests that the observed transition may be correlated directly to a change in the motion of the aromatic phenyl ring. Qualitatively, the fact that a methyl group was removed, while the unit cell remains (virtually) the same, seems consistent with the increased mobility of the phenyl ring (the unit cell is actually slightly larger for N-f-MLF-OH). The measured temperature dependence of the Phe ring flips (based on the static 2H measurements) indicates a drastic reduction of the flipping rate from ~ 105 s−1 near room temperature to <100 s−1 when the temperature drops below 190K. In this regime of slow motion, it is conceivable that the ring can become kinetically trapped in an intermediate orientation. Because of packing interactions in the lattice, the ring flip can be accommodated through concerted motions of surrounding side chains, resulting in the long-range conformational changes evidenced, in turn, by changes in the chemical shift and quantitative structural and dynamical solid state NMR measurements. Changes in the proximity and orientation of the phenyl ring can also directly (and substantially) change the chemical shifts of surrounding nuclei through ring current effects 77. This complicates the interpretation of chemical shift changes in terms of local structural changes, and highlights the importance of explicit quantification of distances, torsion angles and dynamics (as shown here). The structural measurements show that the predominant structural changes are in the χ2 side chain torsion angles of the Met and Leu residues, which are close to the Phe ring (see Figure 8) and thus could correlate to its change in dynamics.

Since no intermediate rate exchange dynamics are observed, it appears that our measurements elucidate the endpoints of the structural and dynamic transition but might not provide direct information on its mechanism. It is reasonable to ask whether or not the activation of these dynamical modes is necessary for the structural phase transition to occur, or if the reverse is true. Further, the experiments presented here do not directly probe the degree to which motion or transient reorganization of the lattice itself is coupled to the dynamics 78 79. Both questions might be probed by molecular dynamics simulations of many unit cells of the N-f-MLF-OH lattice. The latter issue can be studied in isolation using NMR samples that are prepared to yield exclusively intermolecular couplings (e.g. 50% 15N- N-f-MLF-OH in 50% 13C- N-f-MLF-OH), or by approaches based on 1H-1H spin diffusion 80. Further studies may also prove important for the accurate estimation of dipolar couplings and torsion angles in the presence of dynamics 81, a condition that is likely to occur in a physiological environment.

Protein dynamics and applications at low temperature

The tripeptide N-f-MLF-OH is a model system that experiences a subset of the structural dynamics experienced by proteins. While methyl dynamics have been invoked to explain the glass transition in the past 13, one increasingly distinguishes the canonical protein glass transition dominated by coupling to the solvent from distinct contributions due to intrinsic polypeptide dynamics. A complete understanding of these low temperature transitions is limited by a lack of site-resolved experimental data on the structure and motions involved. We have shown that low temperature MAS SSNMR can provide site specific insights into the molecular nature of low temperature dynamics in peptides and proteins. In general, low-temperature applications of solid state NMR include not only conventional NMR on microcrystalline model proteins 82,83, but also dynamic nuclear polarization experiments in membrane proteins and other systems39, 84, 85. Examining the source of line broadening sometimes reported at reduced temperatures is therefore also of practical interest. We observe here that line broadening as a result of cooling to low temperatures (even in a dynamic system) does not preclude low temperature NMR experiments, but that apparent line broadening in these spectra reflects discrete forms of the peptide that are resolved in spectra of higher dimensionality. Remarkably, we actually observe a reduction in this multiplicity at even lower temperatures (near 90K). It will be of interest to examine whether this type of behavior is unique to this system, or can also be observed in other peptides as well as in hydrated proteins, where solvent-dependent mechanisms may result in broadening of NMR lines near the protein glass transition. Indeed, the observation that temperature-induced line broadening is reversible in microcrystalline systems tends to exclude cold-denaturation or damage to the crystalline lattice by freezing as a source of this broadening82, 83. Line broadening and intensity losses can also result from motional interference with heteronuclear 1H-13C and 1H-15N decoupling37,38, as is observed for the Phe ring in N-f-MLF-OH at room temperature and methyl groups at reduced temperatures. Figure 2 shows that deuteration of side chains can serve as one approach to circumvent this problem. In a complicated system such as a hydrated protein, we expect many such mechanisms to contribute to the observed behavior at low temperatures. In our solvent-free system, we can show that the dynamics of the aromatic ring appear correlated with temperature-driven structural transitions in this peptide. In general, our results suggest that solid state NMR can make an important contribution to studies of low temperature dynamics of proteins. This can extend to larger protein systems, but due to the complexities involved we feel that simulations and measurements on model systems such as the one described here will prove essential for a thorough biophysical understanding. Experimental insights such as those presented here can serve as checks of, and input for, detailed computational analyses. We intend to obtain further structural and dynamical data on the peptide crystal lattices discussed here in ongoing solid state NMR studies.

Supplementary Material

1_si_001. Supporting Information Available.

Supporting information includes: schematic illustrations of employed pulse sequences, tables of chemical shifts and structural data obtained at various temperatures, crystal powder diffraction data, and figure showing the temperature-dependence of the 1H-13C dipolar couplings.

ACKNOWLEDGMENTS

We thank Dr. Lisa Tucker-Kellogg for discussions and help with the structure calculations, Ajay Thakkar, Jeff Bryant, and Dr. David Ruben for valuable technical assistance, and Scott Speakman for performing the x-ray powder diffraction measurements. Figure 13 was produced using the UCSF Chimera package from the Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco (supported by NIH P41 RR-01081). This work was supported by grants from the National Institutes of Health (EB-001960, EB-003151 and EB-002026). VSB acknowledges the receipt of a PGS Fellowship from the Natural Sciences and Engineering Research Council of Canada.

References

  • 1.Kendrew JC, Bodo G, Dintzis HM, Parrish RG, Wyckoff H, Phillips DC. A Three-Dimensional Model of the Myoglobin Molecule Obtained by X-Ray Analysis. Nature. 1958;181(4610):662–666. doi: 10.1038/181662a0. [DOI] [PubMed] [Google Scholar]
  • 2.Elber R, Karplus M. Multiple Conformational States of Proteins - a Molecular-Dynamics Analysis of Myoglobin. Science. 1987;235(4786):318–321. doi: 10.1126/science.3798113. [DOI] [PubMed] [Google Scholar]
  • 3.Elber R, Karplus M. Low-Frequency Modes in Proteins - Use of the Effective-Medium Approximation to Interpret the Fractal Dimension Observed in Electron-Spin Relaxation Measurements. Physical Review Letters. 1986;56(4):394–397. doi: 10.1103/PhysRevLett.56.394. [DOI] [PubMed] [Google Scholar]
  • 4.Kushida T, Ahn JS, Hirata K, Kurita A. Glass transition of deoxymyoglobin probed by optical absorption spectroscopy. Biochemical and biophysical research communications. 1989;160(2):948–953. doi: 10.1016/0006-291x(89)92527-8. [DOI] [PubMed] [Google Scholar]
  • 5.Rasmussen BF, Stock AM, Ringe D, Petsko GA. Crystalline ribonuclease A loses function below the dynamical transition at 220 K. Nature. 1992;357(6377):423–424. doi: 10.1038/357423a0. [DOI] [PubMed] [Google Scholar]
  • 6.Lichtenegger H, Doster W, Kleinert T, Birk A, Sepiol B, Vogl G. Heme-Solvent Coupling: A Mossbauer Study of Myoglobin in Sucrose. Biophys. J. 1999;76(1):414–422. doi: 10.1016/S0006-3495(99)77208-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Ding XC, Rasmussen BF, Petsko GA, Ringe D. Direct crystallographic observation of an acyl-enzyme intermediate in the elastase-catalyzed hydrolysis of a peptidyl ester substrate: Exploiting the “glass transition” in protein dynamics. Bioorganic Chemistry. 2006;34(6):410–423. doi: 10.1016/j.bioorg.2006.10.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Pieper J, Irrgang KD, Ratsep M, Voigt J, Renger G, Small GJ. Assignment of the lowest Q(y)-state and spectral dynamics of the CP29 chlorophyll a/b antenna complex of green plants: A hole-burning study. Photochemistry and Photobiology. 2000;71(5):574–581. doi: 10.1562/0031-8655(2000)071<0574:aotlqy>2.0.co;2. [DOI] [PubMed] [Google Scholar]
  • 9.Lechner RE, Fitter J, Dencher NA, Hau T. Low-energy dynamics and biological function. Physica B: Condensed Matter. 2006;385–386(Part 2):835–837. [Google Scholar]
  • 10.Iben IET, Braunstein D, Doster W, Frauenfelder H, Hong MK, Johnson JB, Luck S, Ormos P, Schulte A, Steinbach PJ, Xie AH, Young RD. Glassy behavior of a protein. Physical Review Letters. 1989;62(16):1916. doi: 10.1103/PhysRevLett.62.1916. [DOI] [PubMed] [Google Scholar]
  • 11.Ringe D, Petsko GA. The ‘glass transition’ in protein dynamics: what it is, why it occurs, and how to exploit it. Biophysical Chemistry. 2003;105:667–680. doi: 10.1016/s0301-4622(03)00096-6. [DOI] [PubMed] [Google Scholar]
  • 12.Vitkup D, Ringe D, Petsko GA, Karplus M. Solvent mobility and the protein 'glass' transition. Nature Structural Biology. 2000;7(1):34–38. doi: 10.1038/71231. [DOI] [PubMed] [Google Scholar]
  • 13.Lee AL, Wand AJ. Microscopic origins of entropy, heat capacity and the glass transition in proteins. Nature. 2001;411(6836):501–504. doi: 10.1038/35078119. [DOI] [PubMed] [Google Scholar]
  • 14.Gutin A, Sali A, Abkevich V, Karplus M, Shakhnovich EI. Temperature dependence of the folding rate in a simple protein model: Search for a "glass" transition. Journal of Chemical Physics. 1998;108(15):6466–6483. [Google Scholar]
  • 15.Mok KH, Nagashima T, Day IJ, Jones JA, Jones CJV, Dobson CM, Hore PJ. Rapid sample-mixing technique for transient NMR and photo-CIDNP spectroscopy: Applications to real-time protein folding. Journal of the American Chemical Society. 2003;125(41):12484–12492. doi: 10.1021/ja036357v. [DOI] [PubMed] [Google Scholar]
  • 16.Caliskan G, Briber RM, Thirumalai D, Garcia-Sakai V, Woodson SA, Sokolov AP. Dynamic transition in tRNA is solvent induced. Journal of the American Chemical Society. 2006;128(1):32–33. doi: 10.1021/ja056444i. [DOI] [PubMed] [Google Scholar]
  • 17.Teeter MM, Yamano A, Stec B, Mohanty U. On the nature of a glassy state of matter in a hydrated protein: Relation to protein function. Proceedings of the National Academy of Sciences of the United States of America. 2001;98(20):11242–11247. doi: 10.1073/pnas.201404398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Rassmussen BF, Stock AM, Wallon G, Lovat S, Oshima T, Kanaoka M, Zavodsky P, Ringe D, Petsko GA. Probing the Relationship between Protein Flexibility and Protein Function through the Glass-Transition in Dynamic Behavior at 220-K. Faseb Journal. 1992;6(1):A411–A411. [Google Scholar]
  • 19.Mayer E. Ftir Spectroscopic Study of the Dynamics of Conformational Substates in Hydrated Carbonyl-Myoglobin Films Via Temperature-Dependence of the Co Stretching Band Parameters. Biophysical Journal. 1994;67(2):862–873. doi: 10.1016/S0006-3495(94)80547-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Cordone L, Ferrand M, Vitrano E, Zaccai G. Harmonic behavior of trehalose-coated carbon-monoxy-myoglobin at high temperature. Biophysical Journal. 1999;76(2):1043–1047. doi: 10.1016/S0006-3495(99)77269-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Tsai AM, Neumann DA, Bell LN. Molecular dynamics of solid-state lysozyme as affected by glycerol and water: A neutron scattering study. Biophysical Journal. 2000;79(5):2728–2732. doi: 10.1016/S0006-3495(00)76511-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Cornicchi E, Marconi M, Onori G, Paciaroni A. Controlling the protein dynamical transition with sugar-based bioprotectant matrices: A neutron scattering study. Biophysical Journal. 2006;91(1):289–297. doi: 10.1529/biophysj.106.081752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Roh JH, Novikov VN, Gregory RB, Curtis JE, Chowdhuri Z, Sokolov AP. Onsets of Anharmonicity in Protein Dynamics. Physical Review Letters. 2005;95(3):038101–038104. doi: 10.1103/PhysRevLett.95.038101. [DOI] [PubMed] [Google Scholar]
  • 24.Pieper J, Hauss T, Buchsteiner A, Baczynski K, Adamiak K, Lechner RE, Renger G. Temperature- and Hydration-Dependent Protein Dynamics in Photosystem II of Green Plants Studied by Quasielastic Neutron Scattering. Biochemistry. 2007;46(40):11398–11409. doi: 10.1021/bi700179s. [DOI] [PubMed] [Google Scholar]
  • 25.Doster W, Bachleitner A, Dunau R, Hiebl M, Luscher E. Thermal-Properties of Water in Myoglobin Crystals and Solutions at Subzero Temperatures. Biophysical Journal. 1986;50(2):213–219. doi: 10.1016/S0006-3495(86)83455-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Boswell PG, Lugert EC, Rabai J, Amin EA, Buhlmann P. Coordinative properties of highly fluorinated solvents with amino and ether groups. Journal of the American Chemical Society. 2005;127(48):16976–16984. doi: 10.1021/ja055816k. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Smith J, Kuczera K, Karplus M. Dynamics of Myoglobin - Comparison of Simulation Results with Neutron-Scattering Spectra. Proceedings of the National Academy of Sciences of the United States of America. 1990;87(4):1601–1605. doi: 10.1073/pnas.87.4.1601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Beshah K, Olejniczak ET, Griffin RG. Deuterium NMR Study of Methyl-Group Dynamics in Polycrystalline L-Alanine. Journal of Chemical Physics. 1987;86(9):4730–4736. [Google Scholar]
  • 29.Beshah K, Griffin RG. Deuterium Quadrupole Echo Nmr-Study of Methyl-Group Dynamics in N-Acetyl-Dl-(Gamma-D6)-Valine. Journal of Magnetic Resonance. 1989;84(2):268–274. [Google Scholar]
  • 30.Gavuzzo E, Mazza F, Pochetti G, Scatturin A. Crystal structure, conformation, and potential energy calculations of the chemotactic peptide N-formyl-L-Met-L-Leu-L-Phe-OMe. Int. J. Peptide Protein Res. 1989;34:409–415. doi: 10.1111/j.1399-3011.1989.tb00710.x. [DOI] [PubMed] [Google Scholar]
  • 31.DeLano WL, editor. The PyMOL Molecular Graphics System. Palo Alto, CA, USA: DeLano Scientific; 2002. [Google Scholar]
  • 32.Hong M, Griffin RG. Resonance assignments for solid peptides by dipolar-mediated C-13/N-15 correlation solid-state NMR. Journal of the American Chemical Society. 1998;120(28):7113–7114. [Google Scholar]
  • 33.Ladizhansky V, Veshtort M, Griffin RG. NMR determination of the torsion angle Psi in alpha-helical peptides and proteins: The HCCN dipolar correlation experiment. Journal of Magnetic Resonance. 2002;154(2):317–324. doi: 10.1006/jmre.2001.2488. [DOI] [PubMed] [Google Scholar]
  • 34.Jaroniec CP, Tounge BA, Herzfeld J, Griffin RG. Frequency selective heteronuclear dipolar recoupling in rotating solids: Accurate C-13-N-15 distance measurements in uniformly C-13,N-15-labeled peptides. Journal of the American Chemical Society. 2001;123(15):3507–3519. doi: 10.1021/ja003266e. [DOI] [PubMed] [Google Scholar]
  • 35.Sun BQ, Rienstra CM, Costa PR, Williamson JR, Griffin RG. 3D 15N-13C-13C chemical shift correlation spectroscopy in rotating solids. J. Am. Chem. Soc. 1997;119:8540–8546. [Google Scholar]
  • 36.Rienstra CM, Tucker-Kellogg L, Jaroniec CP, Hohwy M, Reif B, McMahon MT, Tidor B, Lozano-Perez T, Griffin RG. De Novo Determination of Peptide Structure with Solid-State MAS NMR. Proc. Nat. Acad. Sci. USA. 2002;99:10260–10265. doi: 10.1073/pnas.152346599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Long JR, Sun BQ, Bowen A, Griffin RG. Molecular dynamics and magic angle spinning NMR. J. Am. Chem. Soc. 1994;116:11950–11956. [Google Scholar]
  • 38.Maus DC, Copie V, Sun BQ, Griffiths JM, Griffin RG, Luo SF, Schrock RR, Liu AH, Seidel SW, Davis WM, Grohmann A. A solid-state NMR study of tungsten methyl group dynamics in [W(eta(5)-C(5)Me(5))Me(4)][PF6] Journal of the American Chemical Society. 1996;118(24):5665–5671. [Google Scholar]
  • 39.Bajaj VS, Farrar CT, Hornstein MK, Mastovsky I, Vieregg J, Bryant J, Elena B, Kreischer KE, Temkin RJ, Griffin RG. Dynamic nuclear polarization at 9T using a novel 250 GHz gyrotron microwave source. J. Mag. Res. 2003;160:85–90. doi: 10.1016/j.jmr.2011.09.010. [DOI] [PubMed] [Google Scholar]
  • 40.Bajaj VS, Hornstein MK, Mak ML, Herzfeld J, Griffin RG. A Magic Angle Spinning NMR Probe for Dynamic Nuclear Polarization Experiments at 9T (in preparation. J. Mag. Res. 2005 [Google Scholar]
  • 41.Jaroniec CP, MacPhee CE, Bajaj VS, McMahon MT, Dobson CM, Griffin RG. High-resolution molecular structure of a peptide in an amyloid fibril determined by magic angle spinning NMR spectroscopy. Proceedings of the National Academy of Sciences of the United States of America. 2004;101(3):711–716. doi: 10.1073/pnas.0304849101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Gullion T, Schaefer J. Rotational-echo double-resonance NMR. J. Magn. Reson. 1989;81:196–200. doi: 10.1016/j.jmr.2011.09.003. [DOI] [PubMed] [Google Scholar]
  • 43.Jaroniec CP, Filip C, Griffin RG. 3D TEDOR NMR experiments for the simultaneous measurement of multiple carbon-nitrogen distances in uniformly 13C, 15N-labeled solids. J. Am. Chem. Soc. 2002;124(36):10728–10742. doi: 10.1021/ja026385y. [DOI] [PubMed] [Google Scholar]
  • 44.Hong M, Gross JD, Griffin RG. Site-resolved determination of peptide torsion angle phi from the relative orientations of backbone N-H and C-H bonds by solid-state NMR. Journal of Physical Chemistry B. 1997;101(30):5869–5874. [Google Scholar]
  • 45.Reif B, Hohwy M, Jaroniec CP, Rienstra CM, Griffin RG. NH-NH vector correlation in peptides by solid-state NMR. Journal of Magnetic Resonance. 2000;145(1):132–141. doi: 10.1006/jmre.2000.2067. [DOI] [PubMed] [Google Scholar]
  • 46.Feng X, Eden M, Brinkmann A, Luthman H, Eriksson L, Gräslund A, Antzutkin ON, Levitt MH. Direct determination of a peptide torsional angle ψ by double-quantum solid-state NMR. J. Am. Chem. Soc. 1997;119:12006–12007. [Google Scholar]
  • 47.Feng X, Lee YK, Sandström D, Eden M, Maisel H, Sebald A, Levitt MH. Direct determination of a molecular torsional angle by solid-state NMR. Chem. Phys. Lett. 1996;257:314–320. [Google Scholar]
  • 48.Hohwy M, Jaroniec CP, Lansing J, Rienstra CM, Griffin RG. 13C-1H recoupling in the HCCH torsion angle experiment in solid-state NMR. Chem. Phys. Lett. 1999 [Google Scholar]
  • 49.Hong M, Gross JD, Hu W, Griffin RG. Determination of the peptide torsion angle phi by N-15 chemical shift and C-13(alpha)-H-1(alpha) dipolar tensor correlation in solid-state MAS NMR. Journal of Magnetic Resonance. 1998;135(1):169–177. doi: 10.1006/jmre.1998.1573. [DOI] [PubMed] [Google Scholar]
  • 50.Hong M, Gross JD, Rienstra CM, Griffin RG, Kumashiro KK, Schmidt-Rohr K. Coupling amplification in 2D MAS NMR and its application to torsion angle determination in peptides. J. Magn. Reson. 1997;129:85–92. doi: 10.1006/jmre.1997.1242. [DOI] [PubMed] [Google Scholar]
  • 51.Ladizhansky V, Jaroniec CP, Diehl A, Oschkinat H, Griffin RG. Measurement of Multiple psi Angles in Uniformly Labeled α-Spectrin SH3 Domain Using 3D 15N-13C-13C-15N MAS Dipolar-Chemical Shift Correlation Spectroscopy. J. Amer. Chem. Soc. 2003;125:6827–6833. doi: 10.1021/ja029082c. [DOI] [PubMed] [Google Scholar]
  • 52.Rienstra CM, Hohwy M, Mueller LJ, Jaroniec CP, Reif B, Griffin RG. Determination of multiple torsion angle-constraints in U-13C, 15N-labeled peptides: 3D 1H-15N-13C-1H dipolar chemical shift spectroscopy in rotating solids. J. Am. Chem. Soc. 2002;124:11908–11922. doi: 10.1021/ja020802p. [DOI] [PubMed] [Google Scholar]
  • 53.Rienstra CM, Hohwy M, Hong M, Griffin RG. 2D and 3D 15N-13C-13C NMR chemical shift correlation spectroscopy of solids: Assignment of MAS spectra of peptides. J. Am. Chem. Soc. 2000;122:10979–10990. [Google Scholar]
  • 54.Morcombe CR, Zilm KW. Chemical shift referencing in MAS solid state NMR. Journal of Magnetic Resonance. 2003;162(2):479–486. doi: 10.1016/s1090-7807(03)00082-x. [DOI] [PubMed] [Google Scholar]
  • 55.Markley JL, Bax A, Arata Y, Hilbers CW, Kaptein R, Sykes BD, Wright PE, Wüthrich K. Recommendations for the presentation of NMR structures of proteins and nucleic acids. Pure & Appl. Chem. 1998;70:117–142. doi: 10.1046/j.1432-1327.1998.2560001.x. [DOI] [PubMed] [Google Scholar]
  • 56.Levitt MH. Symmetry-Based Pulse Sequences in Magic Angle Spinning Solid State NMR. In: Harris DMGaRK., editor. Encyclopedia of Nuclear Magnetic Resonance. Vol. 9. John Wiley & Sons; 2002. pp. 165–196. [Google Scholar]
  • 57.Delaglio F, Grzesiek S, Vuister GW, Pfeifer J, Bax A. NMR PIPE. J. Biomol. NMR. 1995;6:277–293. doi: 10.1007/BF00197809. [DOI] [PubMed] [Google Scholar]
  • 58.Goddard TD, Kneller DG. SPARKY - NMR Assignment Program. 2006. http://www.cgl.ucsf.edu/home/sparky/) [Google Scholar]
  • 59.Fogh RH, Vranken WF, Boucher W, Stevens TJ, Laue ED. A nomenclature and data model to describe NMR experiments. Journal of Biomolecular Nmr. 2006;36(3):147–155. doi: 10.1007/s10858-006-9076-z. [DOI] [PubMed] [Google Scholar]
  • 60.Vranken WF, Boucher W, Stevens TJ, Fogh RH, Pajon A, Llinas P, Ulrich EL, Markley JL, Ionides J, Laue ED. The CCPN data model for NMR spectroscopy: Development of a software pipeline. Proteins-Structure Function and Bioinformatics. 2005;59(4):687–696. doi: 10.1002/prot.20449. [DOI] [PubMed] [Google Scholar]
  • 61.Fogh RH, Boucher W, Vranken WF, Pajon A, Stevens TJ, Bhat TN, Westbrook J, Ionides JMC, Laue ED. A framework for scientific data modeling and automated software development. Bioinformatics. 2005;21(8):1678–1684. doi: 10.1093/bioinformatics/bti234. [DOI] [PubMed] [Google Scholar]
  • 62.Pajon A, Ionides J, Diprose J, Fillon J, Fogh R, Ashton AW, Berman H, Boucher W, Cygler M, Deleury E, Esnouf R, Janin J, Kim R, Krimm I, Lawson CL, Oeuillet E, Poupon A, Raymond S, Stevens T, van Tilbeurgh H, Westbrook J, Wood P, Ulrich E, Vranken W, Li XL, Laue E, Stuart DI, Henrick K. Design of a data model for developing laboratory information management and analysis systems for protein production. Proteins-Structure Function and Bioinformatics. 2005;58(2):278–284. doi: 10.1002/prot.20303. [DOI] [PubMed] [Google Scholar]
  • 63.Davis JH, Jeffrey KR, Bloom M, Valic MI, Higgs TP. Quadrupolar echo deuteron magnetic resonance spectroscopy in ordered hydrocarbon chains. Chem. Phys. Lett. 1976;42:390–394. [Google Scholar]
  • 64.Wittebort RJ, Olejniczak ET, Griffin RG. Analysis of Deuterium Nuclear-Magnetic-Resonance Line-Shapes in Anisotropic Media. Journal of Chemical Physics. 1987;86(10):5411–5420. [Google Scholar]
  • 65.Hohwy M, Jaroniec CP, Reif B, Rienstra CM, Griffin RG. Local structure and relaxation in solid-state NMR: Accurate measurement of amide N-H bond lengths and H-N-H bond angles. Journal of the American Chemical Society. 2000;122(13):3218–3219. [Google Scholar]
  • 66.Feng X, Verdegem P, Lee Y, Sandstrom D, Eden M, Bovee-Geurts P, de Grip W, Lugtenburg J, de Groot H, Levitt M. Direct Determination of a Molecular Torsion Angle in the Membrane Protein Rhodopsin by Solid-State NMR. J. Am. Chem. Soc. 1997;119(29):6853–6857. [Google Scholar]
  • 67.Hohwy M, Rienstra CM, Griffin RG. Band-selective homonuclear dipolar recoupling in rotating solids. Journal of Chemical Physics. 2002;117(10):4973–4987. [Google Scholar]
  • 68.Costa PR, Gross JD, Hong M, Griffin RG. Solid-state NMR Measurement of ψ in Peptides: a NCCN 2Q-Heteronuclear Local Field Experiment. Chem. Phys. Lett. 1997;280:95–103. [Google Scholar]
  • 69.Takegoshi K, Nakamura S, Terao T. C-13-H-1 dipolar-assisted rotational resonance in magic-angle spinning NMR. Chemical Physics Letters. 2001;344(5–6):631–637. [Google Scholar]
  • 70.Takegoshi K, Nakamura S, Terao T. J. Chem. Phys. 2003;118:2325–2341. [Google Scholar]
  • 71.Morcombe CR, Gaponenko V, Byrd RA, Zilm KW. J. Am. Chem. Soc. 2004;126:7196–7197. doi: 10.1021/ja047919t. [DOI] [PubMed] [Google Scholar]
  • 72.Rice DM, Meinwald YC, Scheraga HA, Griffin RG. Tyrosyl Motion in Peptides - H-2 Nmr Line-Shapes and Spin- Lattice Relaxation. Journal of the American Chemical Society. 1987;109(6):1636–1640. [Google Scholar]
  • 73.Kamihira M, Naito A, Tuzi S, Saito H. Phenyl Ring Dynamics of Enkephalin Molecules and Behavior of Bound Solvents in the Crystalline States by 2H NMR Spectroscopy. J. Phys. Chem. A. 1999;103(18):3356–3363. [Google Scholar]
  • 74.Schwieters CD, Clore GM. Internal coordinates for molecular dynamics and minimization in structure determination and refinement. Journal of Magnetic Resonance. 2001;152(2):288–302. doi: 10.1006/jmre.2001.2413. [DOI] [PubMed] [Google Scholar]
  • 75.Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE. UCSF Chimera - A Visualization System for Exploratory Research and Analysis. J. Comput. Chem. 2004;25(13):1605–1612. doi: 10.1002/jcc.20084. [DOI] [PubMed] [Google Scholar]
  • 76.Cottone G, Giuffrida S, Ciccotti G, Cordone L. Molecular dynamics simulation of sucrose-and trehalose-coated carboxy-myoglobin. Proteins-Structure Function and Bioinformatics. 2005;59(2):291–302. doi: 10.1002/prot.20414. [DOI] [PubMed] [Google Scholar]
  • 77.Ma Z, Halling MD, Solum MS, Harper JK, Orendt AM, Facelli JC, Pugmire RJ, Grant DM, Amick AW, Scott LT. Ring Current Effects in Crystals. Evidence from 13C Chemical Shift Tensors for Intermolecular Shielding in 4,7-Di-t-butylacenaphthene versus 4,7-Di-t-butylacenaphthylene. J. Phys. Chem. A. 2007;111(10):2020–2027. doi: 10.1021/jp068400h. [DOI] [PubMed] [Google Scholar]
  • 78.Pschorn U, Rossler E, Sillescu H, Kaufmann S, Schaefer D, Spiess HW. Local and Cooperative Motions at the Glass-Transition of Polystyrene - Information from One-Dimensional and 2-Dimensional Nmr as Compared with Other Techniques. Macromolecules. 1991;24(2):398–402. [Google Scholar]
  • 79.Schaefer J, Sefcik MD, Stejskal EO, Mckay RA, Dixon WT, Cais RE. Molecular-Motion in Glassy Polystyrenes. Macromolecules. 1984;17(6):1107–1118. [Google Scholar]
  • 80.Elena B, Pintacuda G, Mifsud N, Emsley L. Molecular structure determination in powders by NMR crystallography from proton spin diffusion. Journal of the American Chemical Society. 2006;128(29):9555–9560. doi: 10.1021/ja062353p. [DOI] [PubMed] [Google Scholar]
  • 81.Lorieau JL, McDermott AE. Conformational flexibility of a microcrystalline globular protein: order parameters by solid-state NMR spectroscopy. J. Am. Chem. Soc. 2006;128(35):11505–11512. doi: 10.1021/ja062443u. [DOI] [PubMed] [Google Scholar]
  • 82.Martin RW, Zilm KW. Preparation of protein nanocrystals and their characterization by solid state NMR. Jour. Magn. Resonance. 2003;164:162–174. doi: 10.1016/s1090-7807(03)00253-2. [DOI] [PubMed] [Google Scholar]
  • 83.Franks WT, Zhou DH, Wylie BJ, Money BG, Graesser DT, Frericks HL, Sahota G, Rienstra CM. Magic-Angle Spinning Solid-State NMR Spectroscopy of the b1 Immunoglobulin Binding Domain of Protein G (GB1): 15N and 13C Chemical Shift Assignments and Conformational Analysis. J. Am Chem. Soc. 2005;127:12291. doi: 10.1021/ja044497e. [DOI] [PubMed] [Google Scholar]
  • 84.Hall DA, Maus DC, Gerfen GJ, Inati SJ, Becerra LR, Dahlquist FW, Griffin RG. Polarization-enhanced NMR spectroscopy of biomolecules in frozen solution. Science. 1997;276(5314):930–932. doi: 10.1126/science.276.5314.930. [DOI] [PubMed] [Google Scholar]
  • 85.van der Wel PCA, Hu K-N, Lewandoswki J, Griffin RG. Dynamic Nuclear Polarization of Amyloidogenic Peptide Nanocrystals: GNNQQNY, a Core Segment of the Yeast Prion Protein Sup35p. J. Am Chem. Soc. 2006;128:10840–10846. doi: 10.1021/ja0626685. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1_si_001. Supporting Information Available.

Supporting information includes: schematic illustrations of employed pulse sequences, tables of chemical shifts and structural data obtained at various temperatures, crystal powder diffraction data, and figure showing the temperature-dependence of the 1H-13C dipolar couplings.

NIHMS91489-supplement-1_si_001.pdf (736.2KB, pdf)

RESOURCES