Dynamics of reassembled thioredoxin studied by magic angle spinning NMR: snapshots from different time scales.

Solid-state NMR spectroscopy can be used to probe internal protein dynamics in the absence of the overall molecular tumbling. In this study, we report (15)N backbone dynamics in differentially enriched 1-73(U-(13)C,(15)N)/74-108(U-(15)N) reassembled thioredoxin on multiple time scales using a series of 2D and 3D MAS NMR experiments probing the backbone amide (15)N longitudinal relaxation, (1)H-(15)N dipolar order parameters, (15)N chemical shift anisotropy (CSA), and signal intensities in the temperature-dependent and (1)H T(2)'-filtered NCA experiments. The spin-lattice relaxation rates R(1) (R(1) = 1/T(1)) were observed in the range from 0.012 to 0.64 s(-1), indicating large site-to-site variations in dynamics on pico- to nanosecond time scales. The (1)H-(15)N dipolar order parameters, , and (15)N CSA anisotropies, delta(sigma), reveal the backbone mobilities in reassembled thioredoxin, as reflected in the average = 0.89 +/- 0.06 and delta(sigma) = 92.3 +/- 5.2 ppm, respectively. From the aggregate of experimental data from different dynamics methods, some degree of correlation between the motions on the different time scales has been suggested. Analysis of the dynamics parameters derived from these solid-state NMR experiments indicates higher mobilities for the residues constituting irregular secondary structure elements than for those located in the alpha-helices and beta-sheets, with no apparent systematic differences in dynamics between the alpha-helical and beta-sheet residues. Remarkably, the dipolar order parameters derived from the solid-state NMR measurements and the corresponding solution NMR generalized order parameters display similar qualitative trends as a function of the residue number. The comparison of the solid-state dynamics parameters to the crystallographic B-factors has identified the contribution of static disorder to the B-factors. The combination of longitudinal relaxation, dipolar order parameter, and CSA line shape analyses employed in this study provides snapshots of dynamics and a new insight on the correlation of these motions on multiple time scales.

Magic angle spinning NMR experiments for structural studies of differentially enriched protein interfaces and protein assemblies.

Protein-protein interactions play vital roles in numerous biological processes. These interactions often result in formation of insoluble and noncrystalline protein assemblies. Solid-state NMR spectroscopy is rapidly emerging as a premier method for structural analysis of such systems. We introduce a family of two-dimensional magic angle spinning (MAS) NMR experiments for structural studies of differentially isotopically enriched protein assemblies. Using 1-73((13)C,(15)N)/74-108((15)N) labeled thioredoxin reassembly, we demonstrate that dipolar dephasing followed by proton-assisted heteronuclear magnetization transfer yields long-range (15)N-(13)C correlations arising exclusively from the interfaces formed by the pair of differentially enriched complementary fragments of thioredoxin. Incorporation of dipolar dephasing into the (15)N proton-driven spin diffusion and into the (1)H-(15)N FSLG-HETCOR sequences permits (1)H and (15)N resonance assignments of the 74-108((15)N) enriched C-terminal fragment of thioredoxin alone. The differential isotopic labeling scheme and the NMR experiments demonstrated here allow for structural analysis of both the interface and each interacting protein. Isotope editing of the magnetization transfers results in spectral simplification, and therefore larger protein assemblies are expected to be amenable to these experiments.

Magic angle spinning NMR spectroscopy of thioredoxin reassemblies.

Differentially isotopically enriched 1-73((13)C,(15)N)/74-108((15)N) and 1-73((15)N)/74-108((13)C,(15)N) Escherichia coli thioredoxin reassemblies prepared by fragment complementation were investigated by high-resolution magic angle spinning solid-state NMR spectroscopy. Nearly complete resonance assignments, secondary and tertiary structure analysis are reported for 1-73((13)C,(15)N)/74-108((15)N) reassembled thioredoxin. Temperature dependence of the dipolar-assisted rotational resonance (DARR) spectra reveals the residues undergoing intermediate timescale motions at temperatures below - 15 degrees C. Analysis of the DARR intensity buildups as a function of mixing time in these reassemblies indicates that at long mixing times medium- and long-range cross-peaks do not experience dipolar truncation, suggesting that isotopic dilution is not required for gaining nontrivial distance restraints for structure calculations.

On the NMR analysis of pKa values in the unfolded state of proteins by extrapolation to zero denaturant.

Detailed knowledge of the pH-dependence in both folded and unfolded states of proteins is essential to understand the role of electrostatics in protein stability. The increasing number of natively disordered proteins constitutes an excellent source for the NMR analysis of pKa values in the unfolded state of proteins. However, the tendency of many natively disordered proteins to aggregate via intermolecular hydrophobic clusters limits their NMR analysis over a wide pH range. To assess whether the pKa values in natively disordered polypeptides can be extrapolated from NMR measurements in the presence of denaturants, the natively disordered backbone of the C-terminal fragment 75 to 105 of Human Thioredoxin was studied. First, assignments using triple resonance experiments were performed to confirm lack of secondary structure. Then the pH-dependence of the amides and carboxylate side chains of Glu residues (Glu88, Glu95, Glu98, and Glu103) in the pH range from 2.0 to 7.0 was monitored using 2D 1H15N HSQC and 3D C(CO)NH experiments, and the behavior of their amides and corresponding carboxyl groups was compared to confirm the absence of nonlocal interactions. Lastly, the effect of increasing dimethyl urea concentration on the pKa values of these Glu residues was monitored. The results indicate that: (i) the dispersion in the pKa of carboxyl groups and the pH midpoints of amides in Glu residues is about 0.5 pH units and 0.6 pH units, respectively; (ii) the backbone amides of the Glu residues exhibit pH midpoints which are within 0.2 pH units from those of their carboxylates; (iii) the addition of denaturant produces upshifts in the pKa values of Glu residues that are nearly independent of their position in the sequence; and (iv) these upshifts show a nonlinear behavior in denaturant concentration, complicating the extrapolation to zero denaturant. Nevertheless, the relative ordering of the pKa values of Glu residues is preserved over the whole range of denaturant concentrations indicating that measurements at high denaturant concentration (e.g. 4 M dimethyl urea) can yield a qualitatively correct ranking of the pKa of these residues in natively disordered proteins whose pH-dependence cannot be monitored directly by NMR.

The pH-dependence of amide chemical shift of Asp/Glu reflects its pKa in intrinsically disordered proteins with only local interactions.

Detailed knowledge of the pH-dependence of ionizable residues in both folded and unfolded states of proteins is essential to understand the role of electrostatics in protein folding and stability. The reassembly of E. coli Thioredoxin (Trx) by complementation of its two disordered fragments (1-37/38-108) provides a folded heterodimer in equilibrium with its unfolded state which, based on circular dichroism and NMR spectroscopy, consists of two unfolded monomers. To gain insight into the role of electrostatics in protein folding and stability, we compared the pH-dependence of the carboxylate sidechain chemical shift of each Asp/Glu against that of its backbone amide chemical shift in the unfolded heterodimer. We monitored via C(CO)NH experiments four Asp and four Glu in fragments 38 to 108 (C37) of Trx in the pH range from 2.0 to 7.0 and compared them with results from (1)H(15)N HSQC experiments [Pujato et al., Biophys. J., 89 (2005) 3293-3302]. The (1)H(15)N HSQC analysis indicates two segments with quite distinct behavior: (A) a segment from Ala57 to Ala108 in which ionizable residues have up to three contiguous neighbors with pH-dependent backbone amide shifts, and (B) a segment of fifteen contiguous pH-dependent backbone amide shifts (Leu42 to Val56) in which two Asp and two Glu are implicated in medium range interactions. In all cases, the titration curves are simple modified sigmoidals from which a pH-midpoint (pH(m)) can be obtained by fitting. In segment A, the pH(m) of a given backbone amide of Asp/Glu mirrors within 0.15 pH-units that of its carboxylate sidechain (i.e., the pK(a)). In contrast, segment B shows significant differences with absolute values of 0.46 and 0.74 pH-units for Asp and Glu, respectively. The dispersion in the pH(m) of the backbone amide of Asp/Glu is also different in the two segments. Segment A shows a dispersion of 0.31 and 0.17 pH-units for Asp and Glu, respectively. Segment B shows a substantially larger dispersion (0.50 and 1.08 pH-units for Asp and Glu, respectively). In both segments, the dispersion in the pH(m) of its backbone amide is larger than in the pK(a) of the carboxylate sidechain (the latter is only 0.17 and 0.52 pH-units for Asp and Glu, respectively). Our results indicate that the pH(m) of the backbone amide chemical shift of Asp/Glu in a disordered polypeptide segment is a good predictor of its pK(a) whenever there are none or few neighboring backbone amides with similar pH-dependence.

Elucidating quantitative stability/flexibility relationships within thioredoxin and its fragments using a distance constraint model.

Numerous quantitative stability/flexibility relationships, within Escherichia coli thioredoxin (Trx) and its fragments are determined using a minimal distance constraint model (DCM). A one-dimensional free energy landscape as a function of global flexibility reveals Trx to fold in a low-barrier two-state process, with a voluminous transition state. Near the folding transition temperature, the native free energy basin is markedly skewed to allow partial unfolded forms. Under native conditions the skewed shape is lost, and the protein forms a compact structure with some flexibility. Predictions on ten Trx fragments are generally consistent with experimental observations that they are disordered, and that complementary fragments reconstitute. A hierarchical unfolding pathway is uncovered using an exhaustive computational procedure of breaking interfacial cross-linking hydrogen bonds that span over a series of fragment dissociations. The unfolding pathway leads to a stable core structure (residues 22-90), predicted to act as a kinetic trap. Direct connection between degree of rigidity within molecular structure and non-additivity of free energy is demonstrated using a thermodynamic cycle involving fragments and their hierarchical unfolding pathway. Additionally, the model provides insight about molecular cooperativity within Trx in its native state, and about intermediate states populating the folding/unfolding pathways. Native state cooperativity correlation plots highlight several flexibly correlated regions, giving insight into the catalytic mechanism that facilitates access to the active site disulfide bond. Residual native cooperativity correlations are present in the core substructure, suggesting that Trx can function when it is partly unfolded. This natively disordered kinetic trap, interpreted as a molten globule, has a wide temperature range of metastability, and it is identified as the "slow intermediate state" observed in kinetic experiments. These computational results are found to be in overall agreement with a large array of experimental data.

A selection for mutants that interfere with folding of Escherichia coli thioredoxin-1 in vivo.

Escherichia coli thioredoxin is normally a cytoplasmic protein involved in the reduction of disulfide bonds. However, thioredoxin can be translocated to the periplasm when it is attached to a cotranslational signal sequence. When exported to the periplasm, it can partially replace the activity of DsbA in promoting the formation of disulfide bonds. In contrast, when thioredoxin is fused to a posttranslational signal sequence, very little of it appears in the periplasm. We propose that this absence of posttranslational export is due to the rapid folding of thioredoxin in the cytoplasm. We sought mutants of thioredoxin that retarded its folding in the cytoplasm, which we accomplished by fusing thioredoxin to a posttranslational signal sequence and selecting for mutants in which thioredoxin was exported to the periplasm, where it could replace DsbA. The collection of mutants obtained represents a limited number of amino acid changes in the protein. In vitro studies on purified mutant proteins show that all but one are defective in the kinetics and thermodynamics of protein folding. We propose that the slower folding of the thioredoxin mutant proteins in the cytoplasm allows their export by a posttranslational pathway. We discuss some implications of this class of mutants for aspects of the folding pathway of thioredoxin and for its mechanism of export. In particular, the finding that a folding mutant that allows protein translocation alters an amino acid at the C terminus of the protein suggests that the degree to which thioredoxin folds during its translation must be severely restricted.

pH dependence of amide chemical shifts in natively disordered polypeptides detects medium-range interactions with ionizable residues.

A growing number of natively disordered proteins undergo a folding/binding process that is essential for their biological function. An interesting question is whether these proteins have incompletely solvated regions that drive the folding/binding process. Although the presence of predominantly hydrophobic buried regions can be easily ascertained by high-sensitivity differential scanning calorimetry analysis, the identification of those residues implicated in the burial requires NMR analysis. We have selected a partially solvated natively disordered fragment of Escherichia coli, thioredoxin, C37 (38-108), for full NMR spectral assignment. The secondary chemical shifts, temperature coefficients, and relaxation rates (R(1) and R(2)) of this fragment indicate the presence of a flexible backbone without a stable hydrogen bond network near neutral pH. (1)H-(15)N heteronuclear single quantum coherence analysis of the pH dependence of amide chemical shifts in fragment C37 within pH 2.0 and 7.0 suggests the presence of interactions between nonionizable residues and the carboxylate groups of four Asp and four Glu residues. The pH midpoints (pH(m)) of the amides in the ionizable residues (Asp or Glu) and, consequently, the shifts in the pH(m) (DeltapH(m)) of these residues with respect to model tetrapeptides, are sequence-dependent; and the nonionizable residues that show pH dependence cluster around the ionizable ones. The same pH dependence has been observed in two fragments: M37 (38-73) and C73 (74-108), ruling out the participation of long-range interactions. Our studies indicate the presence of a 15-residue pH-dependent segment with the highest density of ionizable sites in the disordered ensembles of fragments C37 and M37. The observed correlations between ionizable and nonionizable residues in this segment suggest the organization of the backbone and side chains through local and medium-range interactions up to nine residues apart, in contrast to only a few interactions in fragment C73. These results agree qualitatively with the predominantly hydrophobic buried surface detected only in fragments C37 and M37 by highly sensitive differential scanning calorimetry analysis. This work offers a sensitive and rapid new tool to obtain clues about local and nonlocal interactions between ionizable and nonionizable residues in the growing family of natively disordered small proteins with full NMR assignments.

Resonance assignments and secondary structure analysis of E. coli thioredoxin by magic angle spinning solid-state NMR spectroscopy.

De novo site-specific 13C and 15N backbone and sidechain resonance assignments are presented for uniformly enriched E. coli thioredoxin, established using two-dimensional homo- and heteronuclear solid-state magic angle spinning NMR correlation spectroscopy. Backbone dihedral angles and secondary structure were derived from the statistical analysis of the secondary chemical shifts, and are in good agreement with solution values for the intact full-length thioredoxin, with the exception of a small number of residues located at the termini of the individual secondary structure elements. A large number of cross-peaks observed in the DARR spectra with long mixing times correspond to the pairs of carbon atoms separated by 4-6 angstroms, suggesting that DARR could be efficiently employed for observation of medium- and long-range correlations. The 108 amino acid residue E. coli thioredoxin is the largest uniformly enriched protein assigned to this degree of completeness by solid-state NMR spectroscopy to date. It is anticipated that with a combination of two-dimensional correlation experiments and high magnetic fields, resonance assignments and secondary structure can be generally derived for other noncrystalline proteins.

Magic angle spinning solid-state NMR spectroscopy for structural studies of protein interfaces. resonance assignments of differentially enriched Escherichia coli thioredoxin reassembled by fragment complementation.

De novo site-specific backbone and side-chain resonance assignments are presented for U-15N(1-73)/U-13C,15N(74-108) reassembly of Escherichia coli thioredoxin by fragment complementation, determined using solid-state magic angle spinning NMR spectroscopy at 17.6 T. Backbone dihedral angles and secondary structure predicted from the statistical analysis of 13C and 15N chemical shifts are in general agreement with solution values for the intact full-length thioredoxin, confirming that the secondary structure is retained in the reassembled complex prepared as a poly(ethylene glycol) precipitate. The differential labeling of complementary thioredoxin fragments introduced in this work is expected to be beneficial for high-resolution structural studies of protein interfaces formed by protein assemblies by solid-state NMR spectroscopy.

Reversal of negative charges on the surface of Escherichia coli thioredoxin: pockets versus protrusions.

Recent studies of proteins with reversed charged residues have demonstrated that electrostatic interactions on the surface can contribute significantly to protein stability. We have used the approach of reversing negatively charged residues using Arg to evaluate the effect of the electrostatics context on the transition temperature (T(m)), the unfolding Gibbs free energy change (DeltaG), and the unfolding enthalpy change (DeltaH). We have reversed negatively charged residues at a pocket (Asp9) and protrusions (Asp10, Asp20, Glu85), all located in interconnecting segments between elements of secondary structure on the surface of Arg73Ala Escherichia coli thioredoxin. DSC measurements indicate that reversal of Asp in a pocket (Asp9Arg/Arg73Ala, DeltaT(m) = -7.3 degrees C) produces a larger effect in thermal stability than reversal at protrusions: Asp10Arg/Arg73Ala, DeltaT(m) = -3.1 degrees C, Asp20Arg/Arg73Ala, DeltaT(m) = 2.0 degrees C, Glu85Arg/Arg73Ala, DeltaT(m) = 3.9 degrees ). The 3D structure of thioredoxin indicates that Asp20 and Glu85 have no nearby charges within 8 A, while Asp9 does not only have Asp10 as sequential neighbor, but it also forms a 5-A long-range ion pair with the solvent-exposed Lys69. Further DSC measurements indicate that neutralization of the individual charges of the ion pair Asp9-Lys69 with nonpolar residues produces a significant decrease in stability in both cases: Asp9Ala/Arg73Ala, DeltaT(m) = -3.7 degrees C, Asp9Met/Arg73Ala, DeltaT(m) = -5.5 degrees C, Lys69Leu/Arg73Ala, DeltaT(m) = -5.1 degrees C. However, thermodynamic analysis shows that reversal or neutralization of Asp9 produces a 9-15% decrease in DeltaH, while both reversal of Asp at protrusions and neutralization of Lys69 produce negligible changes. These results correlate well with the NMR analysis, which demonstrates that only the substitution of Asp9 produces extensive conformational changes and these changes occur in the surroundings of Lys69. Our results led us to suggest that reversal of a negative charge at a pocket has a larger effect on stability than a similar reversal at a protrusion and that this difference arises largely from short-range interactions with polar groups within the pocket, rather than long-range interactions with solvent-exposed charged groups.

Reduced spectral density mapping of a partially folded fragment of E. coli thioredoxin.

The backbone dynamics of a partially folded, N-terminal fragment of E. coli thioredoxin were investigated using nuclear magnetic resonance spectroscopy (NMR). Relaxation data were collected at three temperatures and analyzed using reduced spectral density mapping. As temperature was increased, the values for the viscosity normalized J(0) and for J(omegaH) increased, while J(omegaN) decreased. The global trend observed for the viscosity normalized J(0) was consistent with an increase in the hydrodynamic volume of the fragment and suggested the presence of correlated rotational motion in the absence of long range interactions. In addition, the residue specific variation observed for the viscosity normalized J(0) suggested contributions to J(omega) from a range of correlation times that are close to the global correlation time.

DSC studies of a family of natively disordered fragments from Escherichia coli thioredoxin: surface burial in intrinsic coils.

The accumulating data from proteome analysis indicates that numerous proteins have segments and/or domains, involved in regulatory functions of the eukaryotic cell, which are entirely unstructured under physiological conditions, challenging the structure-function paradigm. Although many such natively unfolded proteins have been structurally analyzed by NMR spectroscopy, little is known about solvent inaccessible surfaces in premolten globules and intrinsic coils. Recent DSC studies of two protein fragments have shown a promising way to estimate the predominantly hydrophobic buried surfaces [Georgescu, R. E., García-Mira, M. M., Tasayco, M. L., and Sánchez-Ruiz, J. M. (2001) Eur. J. Biochem. 268, 1-10]. Here we report a systematic heat capacity analysis of a family of natively disordered complementary fragments of oxidized Escherichia coli thioredoxin (1-31/32-108, 1-37/38-108, 1-50/51-108, and 38-73) which provides insights into the local and nonlocal interactions contributing to the burial of predominantly hydrophobic surface in intrinsic coils.