Protein Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Alexandrescu, A. T.
Right arrow Articles by Kammerer, R. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Alexandrescu, A. T.
Right arrow Articles by Kammerer, R. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
Protein Science (2003), 12:2132-2140.
Copyright © 2003 The Protein Society

Structure and disorder in the ribonuclease S-peptide probed by NMR residual dipolar couplings

Andrei T. Alexandrescu1 and Richard A. Kammerer2

1 Department of Molecular and Cell Biology, University of Connecticut, Storrs, CT 06269-3125, USA
2 Wellcome Trust Centre for Cell-Matrix Research, University of Manchester, Manchester M13 9PT, UK

Reprint requests to: Andrei Alexandrescu, Department of Molecular and Cell Biology, University of Connecticut, 95 North Eagleville Road, U-3125, Storrs, CT 06269-3125, USA; e-mail: andrei{at}uconn.edu; fax: (860) 486-4331.

(RECEIVED April 28, 2003; FINAL REVISION July 2, 2003; ACCEPTED July 4, 2003)

Supplemental material: See www.proteinscience.org

Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.03164403.


   Abstract
TOP
 Abstract
Introduction
 Results and Discussion
 Materials and methods
 Electronic supplemental material
 References
 
NMR residual dipolar couplings for the S-peptide of ribonuclease A aligned in C8E5/n-octanol liquid crystals are consistent with the presence of a native-like {alpha}-helix structure undergoing dynamic fraying. Residues 3–13, which correspond to the first {alpha}-helix of ribonuclease A, show couplings that become more negative at low temperature and in the presence of salt, conditions which stabilize {alpha}-helical structure in the S-peptide. By contrast, dipolar couplings from the N and C termini of the peptide are close to zero and remain nearly invariant with changes in solution conditions. Torsion angle dynamics simulations using a gradient of dihedral restraint bounds that increase from the center to the ends of the peptide reproduce the experimentally observed sequence dependence of dipolar couplings. The magnitudes of residual dipolar couplings depend on the anisotropy of the solute. Native proteins often achieve nearly spherical shapes due to the hydrophobic effect. Embryonic partially folded structures such as the S-peptide {alpha}-helix have an intrinsically greater potential for anisotropy that can result in sizable residual dipolar couplings in the absence of long-range structure.

Keywords: Protein folding; denatured state; alignment in liquid crystals; protein dynamics; RDC

Abbreviations: C8E5, polyoxyethylene 5 octyl ether • RDC, residual dipolar coupling • RNAseA, ribonuclease A • rms, root mean square • rmsd, root mean square deviation • S-peptide, the 1–20 residue fragment of ribonuclease A


   Introduction
TOP
 Abstract
 Introduction
Results and Discussion
 Materials and methods
 Electronic supplemental material
 References
 
Residual dipolar couplings (RDCs) obtained from solutes oriented in a magnetic field have become increasingly important probes of molecular structure and dynamics (Tjandra and Bax 1997; Prestegard 1998). Applications of dipolar couplings include orientation restraints to improve the accuracy of NMR structures (Tjandra and Bax 1997), structure determination for proteins not amenable to standard NOE distance-restraint methodology (Choy et al. 2001), investigations of orientation dynamics such as hinge movements between domains (Chou et al. 2001), and the characterization of ligands in bound conformations through fast-exchange mediated transfer of dipolar couplings (Koenig et al. 2002). More recently, proteins immersed in the anisotopic environments of liquid crystals or gels have been shown to retain a net sterically induced magnetic alignment even under extreme denaturing conditions (Shortle and Ackerman 2001). Because RDCs of denatured proteins do not correlate with those of the corresponding native states (Shortle and Ackerman 2001; Ackerman and Shortle 2002), their structural basis is poorly understood. At a fundamental level, the structural information encoded in RDCs of denatured proteins impacts models of protein folding. The observation of distinct orientations in denatured proteins precludes a random conformational ensemble, which is often the conceptual starting point in mechanistic models of protein folding. Structure present in the initial states of a folding reaction can promote or hinder progression to the functional native state, as well as affect the thermodynamics of the folding equilibrium (Shortle 1996). At a practical level, studies of denatured proteins can provide a baseline for the interpretation of RDCs in the presence of dynamic disorder.

The S-peptide of ribonuclease A (RNAseA) affords a relatively simple and extensively characterized model system for investigating RDCs of a nascent structure. The S-peptide corresponds to residues 1–20 of RNAseA, and encompasses the first {alpha}-helix of the protein: residues 3–13. Early circular dichroism and NMR chemical shift data showed that {alpha}-helical structure in the S-peptide is stabilized to a level of 30–40%, at temperatures near 0°C in the presence of 0.1 M NaCl (Brown and Klee 1969; Bierzynski et al. 1982; Kim and Baldwin 1984; Nelson and Kallenbach 1989). At temperatures above 25°C, circular dichroism spectra were consistent with an essentially unfolded conformation. The presence of distinct "stop signals," which define the same boundaries for the {alpha}-helix in the S-peptide as in native RNAseA, were supported by NMR chemical shift data (Kim and Baldwin 1984; Nelson and Kallenbach 1989). A recent long-range HNCO investigation of hydrogen bonding in the S-peptide (Jaravine et al. 2001) confirmed that the {alpha}-helix hydrogen bonds stop at the native boundary of residue 13, even at 90% (v/v) TFE. Weak 3JNC' hydrogen bond couplings were detected upstream of the native N terminus Thr 3, however, at TFE concentrations above 30% (v/v). The hydrogen-bond couplings of the S-peptide described a bell-shaped sequence profile, which precludes a uniformly stabilized {alpha}-helix. Hydrogen bond populations were largest at the center of the peptide and decreased gradually towards the ends (Jaravine et al. 2001). 15N relaxation data in the absence of TFE gave a similar profile, consistent with dynamic fraying of the ends of the S-peptide (Alexandrescu et al. 1998).


   Results and Discussion
TOP
 Abstract
 Introduction
 Results and Discussion
Materials and methods
 Electronic supplemental material
 References
 
Temperature and salt dependence of dipolar couplings
Liquid crystals formed from polyoxyethylene 5 octyl ether and 1-octanol (Rückert and Otting 2000) were used to align the S-peptide (5.2% C8E5; r = 1.03) and the complex formed between the S-peptide and the ribonuclease S-protein (4.9% C8E5; r = 1.09). A pH value of 3.8 was used for all samples, as this corresponds to the optimum for {alpha}-helix stability in the free S-peptide (Bierzynski and Baldwin 1982). Figure 1Go shows representative 1JHN data for the S-peptide in the absence of salt at a temperature of 0°C in isotropic (Fig. 1AGo) and oriented solutions (Fig. 1BGo). Figure 2AGo shows 1H-15N RDC data for the S-peptide at a temperature of 0°C as a function of increasing NaCl concentration. Figure 2BGo shows the temperature dependence of RDCs for the S-peptide at a fixed salt concentration of 0.1 M NaCl. In the absence of salt, the RDCs of the free S-peptide ranged from 7 to -7 Hz (Fig. 1Go). Residues 4–10 gave exclusively negative couplings between -1.4 and -7.1 Hz. Addition of 0.1 M salt, which stabilizes {alpha}-helix structure in the free peptide (Brown and Klee 1969; Bierzynski and Baldwin 1982), shifted the RDCs for this region to between -8 and -14 Hz (Table 1Go). The RDCs for residues 4–12 continued to become more negative on further addition of salt, but the largest changes were between 0 and 0.1 M NaCl (Fig. 2AGo). At a fixed salt concentration of 0.1 M NaCl, the RDCs for residues 4–12 decreased by 30% on increasing the temperature from 0 to 5°C, and by 70% from 0 to 21°C (Fig. 2BGo). The small RDCs at the N and C termini (residues -1 to 3 and 13 to 20) were nearly invariant with changes in temperature and salt (Fig. 2Go).



View larger version (34K):
[in this window]
[in a new window]
  		Figure 1. Representative data for the S-peptide at 0°C and pH 3.8, in the absence of salt. Spectra were obtained by addition of experiments that select the bottom right (dark gray) and top right (light gray) components of 1H-15N quartets, along the better-dispersed 15N dimension. The spectrum in (A) was obtained in the absence of liquid crystals. The spectrum in (B) was obtained for the peptide oriented in 5.2% C8E5/n-octanol (r = 1.03). The S-peptide used for NMR studies contains the extra amino acids GS at its N terminus (Alexandrescu et al. 1998). The first residue is not observed due to rapid amide exchange; the second is assigned number "0" so that residues 1–20 in the peptide correspond to residues 1–20 in the RNAseA sequence.

 


View larger version (27K):
[in this window]
[in a new window]
  		Figure 2. 1H-15N RDCs for the S-peptide as a function of salt concentration (A) and temperature (B). The C8E5/n-octanol liquid crystals are stable in the temperature range between -2°C and 24°C and, because they are uncharged, are relatively insensitive to salt concentration (Rückert and Otting 2000). The integrity of the liquid crystals was verified by measuring the 2H splitting of the solvent, which was maintained in the range between 39 and 44 Hz at 500 MHz.

 

View this table:
[in this window]
[in a new window]
  		Table 1. RDCs for the free and bound S-peptide, and Q-factors for fits of the RDC data to residues 3–13 of the ribonuclease A structure
 
Analysis of couplings in terms of the structure of the first {alpha}-helix of RNAseA
The agreement between RDC data and a structural model can be expressed in terms of the Q-factor:


(1)

The Q-factor tends to zero as the agreement between experimental RDCs (RDCexp) and those predicted from a structural model (RDCcalc) improves (Cornilescu et al. 1998; Ottiger and Bax 1999). Figure 3AGo shows the best fit of the RDC data for residues 3–13 in the free S-peptide (0°C, 0.5 M NaCl) to the first 20 residues of the 1.3 Å-resolution X-ray structure of RNAseA (PDB code 7RSA [PDB] ; Wlodawer et al. 1988). The portion of the peptide that adopts {alpha}-helical structure in the native protein gave a Q-factor of 0.19 (dark circles). The Q-factor increased to 0.42 on inclusion of residues 14 and 15, and to 0.50 when all residues of the peptide were included in the fit. The alignment tensor described by residues 3–13 predicted larger absolute RDC values for residues 14–20 than observed experimentally. Figure 3BGo shows the fit obtained for the S-peptide bound to the S-protein (RNAseA residues 21–124). The S-peptide and S-protein form a tight one-to-one complex (Kd < 10-7 M), with a structure and enzymatic activity very similar to that of the wild-type RNAseA (Kim et al. 1992). RDCs for residues 3–13 in the bound S-peptide gave a Q of 0.17 to the corresponding segment in the X-ray structure (Table 1Go). All but the C-terminal residue of the bound peptide agreed with the wild-type RNAseA structure (Q = 0.23 for residues 2–19, Q = 0.43 for residues 2–20).



View larger version (24K):
[in this window]
[in a new window]
  		Figure 3. Fits of the RDC data for the free (A) and the bound (B) S-peptide to the 1.3 Å-resolution X-ray structure of RNAseA. Error bars of ±1 Hz are used to indicate experimentally determined RDC values. Black circles denote back-calculated values from the best-fitting alignment tensor for segment 3–13. The fits gave values of A|| = -8.4 Hz, A{propto} = -2.0 Hz for the free peptide, A|| = -7.9 Hz, and A{propto} = -1.3 Hz for the bound peptide. Gray circles indicate back-calculated values for residues not included in the fit, obtained by assuming the same alignment tensor as for residues 3–13. Fits were performed with MODULE 1.0 (Dosset et al. 2001).

 
The large temperature and salt dependences of RDCs from the free S-peptide (Fig. 2Go) could be due to changes in the structure of the peptide or to changes in the liquid crystal orienting medium. Among the factors that argue against a change in the liquid crystal is that the 2H splitting of the solvent remains relatively constant within 10% (44.0–39.3 Hz) over the range of conditions investigated. By contrast, RDCs from the center of the S-peptide decrease by ~70% when the temperature is increased from 0 to 21°C (Fig. 1BGo). The strongest evidence that the changes in RDCs reflect changes in structure is that the Q-factors (equation 1Go), which are corrected for changes in the alignment tensor, decrease as conditions are shifted towards those that favor native-like {alpha}-helix structure in the S-peptide (Table 1Go). The shape of the RDC sequence profile remains conserved in spite of large changes in Q-factors with changing conditions. The data for the S-peptide are thus correlated between 0.5 M salt and 0°C, where the RDCs reach their maximum negative values, and 0.1 M salt and 21°C, where RDCs are the smallest. In spite of the strong correlation (R = 0.86), the Q-factor for residues 3–13 in the free S-peptide increases from 0.19 to 0.84 between the two sets of conditions (Table 1Go). The correlations are dominated by the largest values from the {alpha}-helix, which show the largest changes with increasing salt. The small RDCs at the chain termini remain nearly constant. The results show that correlations between RDC data sets can occur even when Q-factors no longer support a structural agreement.

Dynamic modulation of RDCs
The dipolar coupling between nuclei i and j is given by


(2)

where rij is the internuclear distance, R = A{perp}/A|| is the rhombicity defined by the ratio of axial (A{perp}) to rhombic (A||) components of the alignment tensor, ({theta},{phi}) are the spherical coordinates describing the orientation of the internuclear vector in the principal axis system of the alignment tensor A, and S is an order parameter that accounts for motional contributions to the dipolar coupling (Tjandra and Bax 1997). S is often taken to be equivalent to the Lipari-Szabo (1982) order parameter (S = SLS). Inclusion of published SLS order parameters for both the free and bound S-peptide (Alexandrescu et al. 1998) made little if any difference in the quality of fits of RDC data to the X-ray structural model. Experimentally determined SLS values (Alexandrescu et al. 1998) ranged from 0.3 to 1.0 for the bound S-peptide with a mean of 0.9, and from 0.3 to 0.8 for the free S-peptide with a mean of 0.7. Setting the order parameters for the free S-peptide to 1.0 increased the Q-factor from 0.45 to 0.50, compared to a global fit in which all residues were assigned experimentally determined SLS values. Other systems, such as the native state of the protein CspA also show very small changes in the quality of RDC fits on inclusion of SLS values (A.T. Alexandrescu, unpubl.). The weak influence of order parameters on RDCs is probably a consequence of the precision with which the two parameters can be measured. Lipari-Szabo order parameters are typically reported as SLS2. A drop from SLS2 = 0.8 to SLS2 = 0.7 corresponds to a change from SLS = 0.89 to SLS = 0.84. Because the experimental uncertainty in RDCs is typically ~1 Hz, a change of ~5% in SLS would only impact RDCs >= 20 Hz. Conversely, for very small SLS2 below 0.2, uncertainties in the order parameter of ~0.05 would result in large errors for RDCs. Whereas SLS2 is sensitive to motions on the nanosecond and subnanosecond timescales of molecular tumbling (Lipari and Szabo 1982), RDCs can be affected by motions on timescales ranging from picoseconds to milliseconds (Prestegard 1998).

Simulations of {alpha}-helix fraying in the S-peptide
The minimum of equation 2Go occurs at {theta} = {phi} = 90°, corresponding to the largest negative couplings, {delta}min {propto} - A||(1 + 1.5R). The maximum of equation 2Go occurs at {theta} = 0°, corresponding to the largest positive couplings, {delta}max {propto} 2A|| (Clore et al. 1998). Equation 2Go reduces to zero for {theta} = 54.74°, {phi} = 45°. The large negative RDCs in the center of the peptide (Fig. 3AGo) predict HN bonds aligned at ~90° with respect to the principal axis, z. The positive couplings, on the other hand, predict an orientation parallel to the principal axis z. The inconsistency of this combination of couplings with {alpha}-helix structure is reflected in the higher Q-factors obtained for the S-peptide at the high temperature and low salt conditions that destabilize {alpha}-helical structure. Moreover, the RDCs near zero suggest N-H bonds orientated with {theta} at the "magic angle" of 54.74° with respect to z. The RDCs of bonds near the magic angle, however, should have the greatest sensitivity to a small change in alignment tensor (Fischer et al. 1999). Together with the large negative couplings at the center of the {alpha}-helix, the nearly constant small RDCs at the ends of the peptide are more consistent with dynamic averaging of couplings to values near zero.

To qualitatively examine the averaging properties of the RDCs, without addressing the time scale of motion, model structures were generated using a torsion angle dynamics/simulated annealing approach commonly used for NMR structure calculations. The calculations included only {phi} and {Psi} dihedral angle restraints, which were obtained from residues 1–20 of the 1.3 Å-resolution structure of RNAseA (Wlodawer et al. 1988). The RDC profile together with previous work (Alexandrescu et al. 1998; Jaravine et al. 2001) clearly suggests that the amplitudes of motions in the free S-peptide vary as a function of position in the sequence (Fig. 3AGo). It would therefore be inappropriate to include a uniform motion correction for all sites. To simulate fraying, dihedral angle restraints from the 7RSA [PDB] X-ray structure were assigned a gradient of "uncertainty" bounds that increased from the center to the ends of the peptide. The {phi} and {Psi} dihedral restraints were thus weighted as: 1(90°), 2(90°), 3(60°), 4(40°), 5(20°), 6(10°), 7(5°), 8(5°), 9(5°), 10(5°), 11(5°), 12(10°), 13(20°), 14(40°), 15(60°), 16(90°), 17(90°), 18(90°), 19(90°), and 20(90°), where the dihedral angle bound for a given position in the amino acid sequence is in parentheses. The ensemble of structures generated by this approach is shown in Figure 5AGo, after superposition on the tightly restrained residues 3–13. The structures were analyzed by two methods. In the first (Fig. 4A,BGo), a least-squares optimization of each structure was performed against the experimental RDC data with the program Module 1.0 (Dosset et al. 2001). In the second (Fig. 4C,DGo), RDCs were simulated from each structure and information on the liquid crystal using the program Pales 2.1 run in "steric mode" (Zweckstetter and Bax 2000).



View larger version (35K):
[in this window]
[in a new window]
  		Figure 5. (A) Superposition of structures used to simulate fraying. Backbone coordinates of residues 3–13 (red) were aligned, giving a best-fit rmsd of 0.8 Å. The N terminus is at the right. (B) Superposition of structures based on rotation matrix-transformed coordinates from steric alignment modeling. Origins represent the centers of rotation for each structure. The poor superposition reflects the spread in alignment tensors, and that HN RDC data obtained for one alignment media give degenerate solutions for 180° rotations about x,y,z (Al-Hashimi et al. 2000). (C) Stereo HN vector-field representation of the ensemble of simulated S-peptide structures aligned as in C. Nitrogen atoms are indicated by red sticks; hydrogen atoms are indicated by spheres colored cyan for residues 4–12, magenta for 3 and 13, and blue for all other residues.

 


View larger version (52K):
[in this window]
[in a new window]
  		Figure 4. Simulations of the effects of dynamic averaging on RDC values. (A) Fits of RDC data for the free S-peptide (0°C, 0.5 M NaCl) to an ensemble of structures generated from {phi},{Psi} dihedral restraints obtained from residues 1–20 in the X-ray structure of RNAseA. To model fraying, restraints were assigned bounds that increased from the center to the ends of the peptide. Back-calculated RDCs from individual structures are indicated by the symbol x. The average over 20 structures is denoted by the filled squares and thick line. (B) Correlation between experimental RDCs and the average of the simulations (dashed line). Correlation between experimental RDCs and the RNAseA X-ray structure (solid line). (C) RDCs predicted from the ensemble of simulated conformations using a Pales "steric" analysis. RDCs for individual structures are indicated with the symbol x. Averages are shown with solid squares. (D) Comparison of experimental RDCs (black circles) with steric Pales predictions for residues 1–20 of the RNAse X-ray structure (the symbol x), and the average of the simulated structures (gray circles). Correlation to experimental values gave R = 0.84 and R = 0.63 for the simulated ensemble and X-ray structure, respectively. Pales results were uniformly scaled down by a factor of 2.9 for the X-ray structure and 2.5 for the average of the simulated structures.

 
Best-fit modeling of dipolar couplings for the ensemble of simulated S-peptide structures
Best fits of experimental RDCs to each member of the ensemble of simulated structures were obtained by optimization of the parameters A||, A{perp}, {alpha}, ß, {gamma} that specify the alignment tensor A (Dosset et al. 2001). The parameters A||, A{perp} depend on the shape of the solute and on the properties of the alignment media (Zweckstetter and Bax 2000). The Euler angles {alpha}, ß, {gamma} specify the orientation of the alignment tensor A, with respect to the coordinate frame of the structure. The rationale for performing fits of the RDC data to an ensemble rather than a single structure is that motifs conserved over the ensemble might reinforce RDCs, whereas structural divergence would result in RDCs of different sign that would interfere destructively. Figure 4AGo shows RDCs back-calculated from the ensemble of simulated structures in Figure 5AGo. The ensemble average (Fig. 4AGo, squares) captures the bell-shaped profile of RDC values observed experimentally and fits the experimental data slightly better than the X-ray structure (Fig. 4BGo). The shift from negative to positive RDCs at the boundaries of the 3–13 helix is a manifestation of fraying. As the HN vectors fall out of alignment with the {alpha}-helix axis oriented along y, the HN vectors tend towards orientations more parallel with the principal axis z. The corresponding RDCs shift from the maximum negative values towards positive values, before the bond orientations become uncorrelated, and the RDCs collapse to zero. The largest positive RDCs for residue 14 and 15 are maintained for all solution conditions examined (Fig. 1Go). Averages over structures with dihedrals restrained to ideal right-handed {alpha}-helix values ({phi} = -57, {Psi} = -47), or ensemble averages of structures in which only the dihedrals of residues in the {alpha}-helix were restrained, gave sequence profiles in which RDCs decayed to zero instead of assuming the distinct positive values observed for residues 14 and 15 (not shown). The conformational preferences of residues 14 and 15 might be related to the stop signal for the C terminus of the {alpha}-helix.

Steric modeling of dipolar couplings for the ensemble of simulated S-peptide structures
As a further test, structures were analyzed using the "prediction of sterically induced alignment" mode of the program Pales 2.1 (Zweckstetter and Bax 2000). Alignment in a neutral liquid crystal is thought to occur through steric restriction of the reorientation of the solute, and thus depends on the shape of the solute (Almond and Axelsen 2002; Azurmendi and Bush 2002). As opposed to a lest-squares fit analysis, the alignment tensor in "steric" simulations is obtained directly by considering solute orientations excluded due to steric clash with the liquid crystal matrix (Zweckstetter and Bax 2000). Pales calculations were run with parameters specifying a liquid crystal wall model (-bic), a liquid crystal concentration of 5.6%, a grid spacing of 0.3 Å, and a default correction factor of 0.8 to account for incomplete alignment of the liquid crystal with the magnetic field. The magnitudes of the predicted RDCs had a strong dependence on the radius assumed for the solute. When only the {alpha}-helix portion of the S-peptide was considered, a rms minimum between experimental and simulated values was obtained with a radius of ~8 Å. This value is close to the distance spanned by residues 3–13 in the X-ray structure of RNAseA. With RDC data for the entire peptide, the rms minimum occurred for a radius of ~70 Å, testifying to the aberrant values of the RDCs at the C terminus of the peptide. As a compromise, the radius was set to 15 Å, a figure corresponding to the radius of gyration of a "random coil" of 20 residues (Creighton 1993). The value chosen for the solute radius scaled the magnitude of the RDCs uniformly, but did not affect the shape of the sequence profile.

In the ensemble of simulated structures, the weakly restrained regions at the ends of the peptide adopt a broad range of conformations (Fig. 5AGo). Because the disordered regions have a comparable length to the well-defined 3–13 {alpha}-helix, the Pales simulations predicted a broad dispersion of rhombicities and alignment tensors (Fig. 5BGo). Nevertheless, as shown in the "HN vector field" diagram of Figure 5CGo, the HN vectors in the {alpha}-helix retain a coherent alignment that reinforces negative RDCs. In contrast, the alignment coherence dissipates for the weakly restrained chain termini, resulting in near-zero average RDCs. The RDC data for the free S-peptide are more consistent with the ensemble of simulated structures (R = 0.84; Fig. 4DGo) than with the single X-ray structure (R = 0.63; Fig. 4DGo), in spite of the fact that the simulations used to model fraying were coarse. Simplifying assumptions included a gradient of restraint bounds symmetrically disposed about the center of the {alpha}-helix. The same bounds were assumed for the {phi} and {Psi} dihedrals of each residue. Deviations of the {omega} angle from planarity, which could introduce additional torsional degrees of freedom, were not considered in the simulations. The principal factor accounting for the better agreement with the ensemble average compared to the X-ray structure is that the RDCs near the ends of the peptide are averaged to values near zero. The X-ray structure, or, for that matter, individual members of the ensemble, predict larger RDCs than observed for the ends of the peptide.

Additional simulations were done with a set of 20 random structures of the same length and sequence as the S-peptide (not shown). The random structures were generated from an unrestrained molecular dynamics run with the program X-PLOR using a published protocol (Nilges et al. 1988). In contrast to the restrained structures, the RDCs predicted from Pales simulations of individual random structures and for the average over the random structures were not correlated with experimental values (R = 0.24). Although the ranges of RDCs predicted for individual random structures were similar to those predicted for the X-ray structure, the average was about fourfold smaller than for the X-ray or the restrained simulated structures. This suggests that dilution of an ensemble of structures with random conformations has the effect of decreasing observed RDCs towards the isotropic limit. The S-peptide has been estimated to contain ~30% {alpha}-helix at 0°C and 1 M NaCl by circular dichroism (Bierzynski and Baldwin 1982). The steric alignment simulations with residues 1–20 from the RNAseA X-ray structure predict dipolar couplings about threefold larger than those observed in the S-peptide, and it is tempting to conclude that this discrepancy is due to the presence of ~70% random conformations. The absolute values of the simulated RDCs, however, are extremely sensitive to the radius assigned to the solute, which for a dynamic system such as the S-peptide is difficult to estimate. Moreover, the orienting power of the liquid crystals is sensitive to the concentrations of the C8E5 and n-octanol components. The measurement of "absolute alignment" to obtain an estimate of the fractional population of unstructured conformers would thus seem fraught with problems. The present work together with previous NMR data (Alexandrescu et al. 1998; Jaravine et al. 2001) show that the stability of the {alpha}-helix varies along the length of the S-peptide. A comparison with a global figure of 30% {alpha}-helix is likely to be ambiguous.

The interpretation of RDCs hinges on the identification of cooperatively oriented domains
The RDCs of a single-domain protein in its native state describe the global anisotropy of a cooperatively folded structure. Because proteins in their functional states are often constrained to nearly spherical shapes, the inherent anisotropy of native proteins is usually small (Tjandra and Bax 1997). In denatured proteins, nascent unfulfilled structures like the S-peptide {alpha}-helix have an intrinsically greater potential for anisotropy. On the other hand, motional disorder in a denatured protein would tend to dampen the magnitudes of RDCs, perhaps accounting for the observation that denatured states are more difficult to align (Ackerman and Shortle 2002). It has recently been proposed that a denatured {Delta}131{Delta} fragment of staphylococcal nuclease has a distinct topology that persists between 0 and 8 M urea (Shortle and Ackerman 2001; Ackerman and Shortle 2002). The conclusion that the fragment has a specific fold was based on the identification of long-range contacts in paramagnetic spin-labeled derivatives of the {Delta}131{Delta} fragment, which allowed a model of the topology of the fragment to be constructed (Gillespie and Shortle 1997). By itself, the observation of sizable dipolar couplings for proteins under extreme denaturing conditions need not imply that the polypetide chain has long-range order, or a globally stabilized topology. A polypeptide chain comprised of two or more independent domains arranged like beads on a string could give sizable dipolar couplings as long as the individual domains are sufficiently anisotropic (Fischer et al. 1999; Onishi and Shortle 2003). Consider, for example, a hypothetical construct consisting of a tandem of two S-peptides, in which the C terminus of the first peptide is linked to the N terminus of the second (see Electronic Supplemental Material). If the two domains were independent, the RDC sequence profile in Figure 2Go would be duplicated, giving two troughs for the two {alpha}-helices and near-zero RDCs for the intervening disordered segment. This same profile, however, could also be consistent with a static structure; for example, an extended chain that has the two {alpha}-helices aligned in parallel. The ability to separate structural from dynamic contributions to dipolar couplings (Tolman et al. 1997; Peti et al. 2002; Tolman 2002) clearly represents an important challenge for future studies of highly dynamic molecules, including denatured proteins.


   Materials and methods
TOP
 Abstract
 Introduction
 Results and Discussion
 Materials and methods
Electronic supplemental material
 References
 
C8E5 (polyoxyethylene 5 Octyl Ether, C18H36O6), 1-octanol (HPLC grade) and ribonuclease S-protein (Grade XII-PR) were from Sigma. The 15N-labled S-peptide was expressed as a fusion protein in Escherichia coli using the pep-Tvector (Brandenberger et al. 1996) and purified as previously described (Alexandrescu et al. 1998). Lyophilized samples of HPLC-purified S-peptide were dissolved in Milli-Q water, and the pH was adjusted to 3.8. To form the liquid crystals, C8E5 was added directly to the sample to a final concentration of 5.2% (v/v). n-Octanol was added in 1 µl increments to a final molar ratio of C8E5/n-octanol of 1.03. The magnetic alignment of the samples was verified from the doublet 2H splitting of the solvent (Rückert and Otting 2000). The salt concentration of samples was adjusted using a 5 M NaCl stock solution. Volume changes used to achieve the highest NaCl concentration of 0.5 M NaCl resulted in a dilution of the C8E5 concentration from 5.2% to 4.7% (the r ratio of 1.03 is unaffected by dilution). A proportional decrease of the 2H quadrupolar splitting of the solvent from 44.0 to 39.3 Hz was observed. For temperature studies, the 2H splitting decreased from 43.3 at a temperature of 0°C, to 41.1 Hz at 21°C. The S-peptide/S-protein complex sample had 1.7 mM S-peptide and 2.9 mM S-protein, and was studied at pH 4.2 and a temperature of 5°C. The liquid crystals used to orient the complex were 4.9% C8E5 (r = 1.06). The 2H splitting of the aligned sample was 40 Hz.

NMR spectra were recorded on a 500-MHz Inova spectrometer. 1JHN couplings were measured from separate 1H-15N correlation experiments, used to select the bottom right TROSY peak (Pervushin et al. 1997) and the top right correlation, of undecoupled 1H-15N quartets along the better-dispersed 15N frequency. The experiments were acquired with spectral widths of 1H 4964 x 15N 900 Hz digitized into 1H 1024 x 15N 128 complex points for the free S-peptide; and 1H 5498 x 15N 1000 Hz digitized into 1H 1024 x 15N 64 complex points for the bound S-peptide. RDCs were obtained from the difference in one-bond couplings between oriented and control samples (RDC = 1JHN aligned - 1JHN unaligned).

For structure calculations, an initial 200–400 structures were generated with the program Dyana 1.5 (Güntert et al. 1997) using dihedral restraints derived from residues 1–20 of the 7RSA [PDB] X-ray structure. The restraint bounds were set as described in the main text. The 20 structures with the smallest Dyana target functions were subjected to further simulated annealing refinement, and energy minimization using the program XPLOR 3.851 (Brünger 1992). A square well potential with a standard force constant of 200 kcal/(mole Å2) was used for the X-PLOR calculations. The structures had no violations from input restraints larger than 2°. Modeling of random structures was done directly in X-PLOR using a published procedure (Nilges et al. 1988). All reported rmsds were calculated over backbone C{alpha}, N, C' atoms.


   Electronic supplemental material
TOP
 Abstract
 Introduction
 Results and Discussion
 Materials and methods
 Electronic supplemental material
References
 
Summary of RDC simulations for a hypothetical tandem molecule formed by linking two S-peptides is available.


   Acknowledgments
 
A.T.A. thanks Dr. Mark Maciejewski for the NMR pulse programs used to measure RDCs. This work was supported by NSF grant MCB 0236316 (to A.T.A.).

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.


   References
TOP
 Abstract
 Introduction
 Results and Discussion
 Materials and methods
 Electronic supplemental material
 References
 
Ackerman, M.S. and Shortle, D. 2002. Molecular alignment of denatured states of staphylococcal nuclease with strained polyacrylamide gels and surfactant liquid crystalline phases. Biochemistry 41: 3089–3095.[CrossRef][Medline]

Alexandrescu, A.T., Rathgeb-Szabo, K., Rumpel, K., Jahnke, W., Schulthess, T., and Kammerer, R.A. 1998. 15N backbone dynamics of the S-peptide from ribonuclease A in its free and S-protein bound forms: Toward a site-specific analysis of entropy changes upon folding. Protein Sci. 7: 389–402.[Abstract]

Al-Hashimi, H.M., Valafar, H., Terrell, M., Zartler, E.R., Eidsness, M.K., and Prestegard, J.H. 2000. Variation of molecular alignment as a means of resolving orientational ambiguities in protein structures from dipolar couplings. J. Magn. Reson. 143: 402–406.[CrossRef][Medline]

Almond, A. and Axelsen, J.B. 2002. Physical interpretation of residual dipolar couplings in neutral aligned media. J. Am. Chem. Soc. 124: 9986–9987.[CrossRef][Medline]

Azurmendi, H.F. and Bush, C.A. 2002. Conformational studies of blood group A and blood group B oligosaccharides using NMR residual dipolar couplings. Carbohydr. Res. 337: 905–915.[CrossRef][Medline]

Bierzynski, A. and Baldwin, R.L. 1982. Local secondary structure in ribonuclease A denatured by guanidine. HCl near 1°C. J. Mol. Biol. 162: 173–186.[CrossRef][Medline]

Bierzynski, A., Kim, P.S., and Baldwin, R.L. 1982. A salt bridge stabilizes the helix formed by isolated C-peptide of RNAseA. Proc. Natl. Acad. Sci. 79: 2470–2474.[Abstract/Free Full Text]

Brandenberger, R., Kammerer, R.A., Engel, J., and Chiquet, M., 1996. Native chick laminin-4 containing the ß 2 chain (s-laminin) promotes motor axon growth. J. Cell Biol. 135: 1583–1592.[Abstract/Free Full Text]

Brown, J.E. and Klee, W.A. 1969. Conformational studies of a series of overlapping peptides from ribonuclease and their relationship to the protein structure. Biochemistry 8: 2876–2879.[CrossRef][Medline]

Brünger, A.T. 1992. X-PLOR Version 3.1: A system for crystallography and NMR. Yale University Press, New Haven.

Chou, J.J., Li, S., Klee, C.B., and Bax, A. 2001. Solution structure of Ca(2+)-calmodulin reveals flexible hand-like properties of its domains. Nat. Struct. Biol. 8: 990–997.[CrossRef][Medline]

Choy, W.Y., Tollinger, M., Mueller, G.A., and Kay, L.E. 2001. Direct structure refinement of high molecular weight proteins against residual dipolar couplings and carbonyl chemical shift changes upon alignment: An application to maltose binding protein. J. Biomol. NMR 21: 31–40.[CrossRef][Medline]

Clore, G.M., Gronenborn, A.M., and Tjandra, N. 1998. Direct structure refinement against residual dipolar couplings in the presence of rhombicity of unknown magnitude. J. Magn. Reson. 131: 159–162.[CrossRef][Medline]

Cornilescu, G., Marquardt, J.L., Ottiger, M., and Bax, A. 1998. Validation of protein structure from anisotropic carbonyl chemical shifts in a dilute liquid crystalline phase. J. Am. Chem. Soc. 120: 6836–6837.[CrossRef]

Creighton, T.E. 1993. Proteins: Structures and molecular properties, 2nd ed. W.H. Freeman and Company, New York.

Dosset, P., Hus, J.C., Marion, D., and Blackledge, M. 2001. A novel interactive tool for rigid-body modeling of multi-domain macromolecules using residual dipolar couplings. J. Biomol. NMR 20: 223–231.[CrossRef][Medline]

Fischer, M.W., Losonczi, J.A., Weaver, J.L., and Prestegard, J.H. 1999. Domain orientation and dynamics in multidomain proteins from residual dipolar couplings. Biochemistry 38: 9013–9022.[CrossRef][Medline]

Gillespie, J.R. and Shortle, D. 1997. Characterization of long-range structure in the denatured state of staphylococcal nuclease. II. Distance restraints from paramagnetic relaxation and calculation of an ensemble of structures. J. Mol. Biol. 268:170–184.[CrossRef][Medline]

Güntert, P., Mumenthaler, C., and Wüthrich, K. 1997. Torsion angle dynamics for NMR structure calculation with the new program DYANA. J. Mol. Biol. 273: 283–298.[CrossRef][Medline]

Jaravine, V.A., Alexandrescu, A.T., and Grzesiek, S. 2001. Observation of the closing of individual hydrogen bonds during TFE-induced helix formation in a peptide. Protein Sci. 10: 943–950.[Abstract/Free Full Text]

Kim, E.E., Varadarajan, R., Wyckoff, H.W., and Richards, F.M. 1992. Refinement of the crystal structure of ribonuclease S. Comparison with and between the various ribonuclease A structures. Biochemistry 31: 12304– 12314.[CrossRef][Medline]

Kim, P.S. and Baldwin, R.L. 1984. A helix stop signal in the isolated S-peptide of ribonuclease A. Nature 307: 329–334.[CrossRef][Medline]

Koenig, B.W., Kontaxis, G., Mitchell, D.C., Louis, J., Litman, B.J., and Bax, A. 2002. Structure and orientation of a G protein fragment in the receptor bound state from residual dipolar couplings. J. Mol. Biol. 322: 441–461.[CrossRef][Medline]

Lipari, G., and Szabo, A. 1982. Model-free approach to the interpretation of nuclear magnetic resonance in macromolecules: 1. Theory and range of validity. J. Am. Chem. Soc. 104: 4546–4559.[CrossRef]

Nelson, J.W., and Kallenbach, N.R. 1989. Persistence of the {alpha}-helix stop signal in the S-peptide in trifluoroethanol solutions. Biochemistry 28: 5256–5261.[CrossRef][Medline]

Nilges, M., Gronenborn, A.M., Brunger, A.T., and Clore, G.M. 1988. Determination of three-dimensional structures of proteins by simulated annealing with interproton distance restraints. Application to crambin, potato carboxypeptidase inhibitor and barley serine proteinase inhibitor 2. Protein Eng. 2: 27–38.[Abstract/Free Full Text]

Onishi, S. and Shortle, D. 2003. Observation of residual dipolar couplings in short peptides. Proteins 50: 546–551.[CrossRef][Medline]

Ottiger, M., and Bax, A. 1999. Bicelle-based liquid crystals for NMR-measurement of dipolar couplings at acidic and basic pH values. J. Biomol. NMR 13: 187–191.[CrossRef][Medline]

Pervushin, K., Riek, R., Wider, G., and Wuthrich, K. 1997. Attenuated T2 relaxation by mutual cancellation of dipole–dipole coupling and chemical shift anisotropy indicates an avenue to NMR structures of very large biological macromolecules in solution. Proc. Natl. Acad. Sci. 94: 12366– 12371.[Abstract/Free Full Text]

Peti, W., Meiler, J., Brüschweiler, R., and Griesinger, C. 2002. Model-free analysis of protein backbone motion from residual dipolar couplings. J. Am. Chem. Soc. 124: 5822–5833.[CrossRef][Medline]

Prestegard, J.H. 1998. New techniques in structural NMR—Anisotropic interactions. Nat. Struct. Biol. Suppl. 5: 17–22.

Rückert, M. and Otting, G. 2000. Alignment of biological macromolecules in novel nonionic liquid crystalline media for NMR experiments. J. Am. Chem. Soc. 122: 7793–7797.[CrossRef]

Shortle, D. 1996. The denatured state (the other half of the folding equation) and its role in protein stability. FASEB J. 10: 27–34.[Abstract]

Shortle, D. and Ackerman, M.S. 2001. Persistence of native-like topology in a denatured protein in 8 M urea. Science 293: 487–489.[Abstract/Free Full Text]

Tjandra, N. and Bax, A. 1997. Direct measurement of distances and angles in biomolecules by NMR in a dilute liquid crystalline medium. Science 278: 1111–1114.[Abstract/Free Full Text]

Tolman, J.R. 2002. A novel approach to the retrieval of structural and dynamic information from residual dipolar couplings using several oriented media in biomolecular NMR spectroscopy. J. Am. Chem. Soc. 124: 12020–12030.[CrossRef][Medline]

Tolman, J.R., Flanagan, J.M., Kennedy, M.A., and Prestegard, J.H. 1997. NMR evidence for slow collective motions in cyanometmyoglobin. Nat. Struct. Biol. 4: 292–297.[CrossRef][Medline]

Wlodawer, A., Svensson, L.A., Sjolin, L., and Gilland, G.L. 1988. Structure of phosphate-free ribonuclease A refined at 1.26 Å. Biochemistry 27: 2705–2717.[CrossRef][Medline]

Zweckstetter, M. and Bax, A. 2000. Predictions of sterically induced alignment in a dilute liquid crystalline phase: Aid to protein structure determination by NMR. J. Am. Chem. Soc. 122: 3791–3792.[CrossRef]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Alexandrescu, A. T.
Right arrow Articles by Kammerer, R. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Alexandrescu, A. T.
Right arrow Articles by Kammerer, R. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS