Automated prediction of 15N, 13Cα, 13Cβ and 13C′ chemical shifts in proteins using a density functional database

A database of peptide chemical shifts, computed at the density functional level, has been used to develop an algorithm for prediction of ¹⁵N and ¹³C shifts in proteins from their structure; the method is incorporated into a program called SHIFTS (version 4.0). The database was built from the calculated chemical shift patterns of 1335 peptides whose backbone torsion angles are limited to areas of the Ramachandran map around helical and sheet configurations. For each tripeptide in these regions of regular secondary structure (which constitute about 40% of residues in globular proteins) SHIFTS also consults the database for information about sidechain torsion angle effects for the residue of interest and for the preceding residue, and estimates hydrogen bonding effects through an empirical formula that is also based on density functional calculations on peptides. The program optionally searches for alternate side-chain torsion angles that could significantly improve agreement between calculated and observed shifts. The application of the program on 20 proteins shows good consistency with experimental data, with correlation coefficients of 0.92, 0.98, 0.99 and 0.90 and r.m.s. deviations of 1.94, 0.97, 1.05, and 1.08 ppm for ¹⁵N, ¹³C, ¹³C and ¹³C, respectively. Reference shifts fit to protein data are in good agreement with `random-coil' values derived from experimental measurements on peptides. This prediction algorithm should be helpful in NMR assignment, crystal and solution structure comparison, and structure refinement.     

TROSY-based HNCO pulse sequences for the measurement of 1HN-15N, 15N-13CO, 1HN-13CO, 13CO-13Cα and 1HN-13Cα dipolar couplings in 15N, 13C, 2H-labeled proteins

HNCO-based 3D pulse schemes are presented for measuring ¹HN-¹⁵N,¹⁵N-¹³CO, ¹HN-¹³CO,¹³CO-¹³C and ¹HN-¹³C dipolar couplings in ¹⁵N,¹³C,²-labeled proteins. The experiments are based on recently developed TROSY methodology for improving spectral resolution and sensitivity. Data sets recorded on a complex of Val, Leu, Ile (1 only) methyl protonated ¹⁵N,¹³C,²H-labeled maltose binding protein and -cyclodextrin as well as ¹⁵N,¹³C,²H-labeled human carbonic anhydrase II demonstrate that precise dipolar couplings can be obtained on proteins in the 30–40 kDa molecular weight range. These couplings will serve as powerful restraints for obtaining global folds of highly deuterated proteins.     

MQ-HCN-based pulse sequences for the measurement of 13C1′-1H1′, 13C1′-15N, 1H1′-15N, 13C1′-13C2′, 1H1′-13C2′, 13C6/8-1H6/8, 13C6/8-15N, 1H6/8-15N, 13C6-13C5, 1H6-13C5 dipolar couplings in 13C, 15N-labeled DNA (and RNA)

A suite of multiple quantum (MQ) HCN-based pulse sequences has been developed for the purpose of collecting dipolar coupling data in labeled nucleic acids. All the pulse sequences are based on the robust MQ-HCN experiment which has been utilized for assignment purposes in labeled nucleic acids for a number of years and provides much-needed resolution for the dipolar coupling measurements. We have attempted to collect multiple couplings centered on the 13C1' and 13C6/8 positions. Six pulse sequences are described, one each for measurement of one-bond 13C1'-1H1' and 13C6/8-1H6/8 couplings, one for measurement of one-bond 13C1'-15N and two-bond 1H1'-15N couplings, one for measurement of one-bond 13C6/8-15N and two-bond 1H6/8-15N couplings, one for measurement of one-bond 13C1'- 13C2' and two-bond 1H1'-13C2' couplings, and one for measurement of one-bond 13C6-13C5 and two-bond 1H6-13C5 couplings in the bases of C and T. These sequences are demonstrated for a labeled 18 bp DNA duplex in a 47 kDa ternary complex of DNA, CBFbeta, and the CBFalpha Runt domain, thus clearly demonstrating the robustness of the pulse sequences even for a very large complex.     

A 4D TROSY-based pulse scheme for correlating 1HNi,15Ni,13Cαi,13C′i−1 chemical shifts in high molecular weight, 15N,13C, 2H labeled proteins

A 4D TROSY-based triple resonance experiment, 4D-HNCO_i–1CA_i, is presented which correlates intra-residue ¹HN, ¹⁵N, ¹³ C chemical shifts with the carbonyl (¹³C) shift of the preceding residue. The experiment is best used in concert with recently described 4D TROSY-HNCOCA and -HNCACO experiments [Yang, D. and Kay, L.E. (1999) J. Am. Chem. Soc., 121, 2571–2575]. In cases where degeneracy of (¹HN,¹⁵N) spin pairs precludes assignment using the HNCOCA and HNCACO, the HNCO_i–1CA_i often allows resolution of the ambiguity by linking the ¹³C and ¹³C spins surrounding the (¹HN,¹⁵N) pair. The experiment is demonstrated on a sample of ¹⁵N, ¹³C, ² H labeled maltose binding protein in complex with -cyclodextrin that tumbles with a correlation time of 46 ns.     

Selective diagonal-free 13C,13C-edited aliphatic–aromatic NOESY experiment with non-uniform sampling

A band-selective aromatic–aliphatic C,C-edited four-dimensional NOESY experiment is proposed here. Its key advantage is the absence of auto-correlation signals which makes it very attractive for joint use with non-uniform sampling. It is demonstrated here that the sensitivity of the experiment is not significantly affected by utilization of selective pulses (for either aromatic-13C or aliphatic-13C spins). The method was applied to the sample of E32Q mutant of human S100A1 protein, a homodimer of total molecular mass ~20 kDa. High-resolution 4D spectra were obtained from ~1.5 % of sampling points required conventionally. It is shown that superior resolution facilitates unambiguous assignment of observed aliphatic–aromatic cross-peaks. Additionally, the addition of aliphatic-13C dimension enables to resolve peaks with degenerated aliphatic ¹H chemical shifts. All observed cross-peaks were validated against previously determined 3D structure of E32Q mutant of S100A1 protein (PDB 2LHL). The increased reliability of structural constraints obtained from the proposed high-resolution 4D 13C(ali),13C(aro)-edited NOESY can be exploited in the automated protocols of structure determination of proteins.     

Determination of HN,Hα and HN,C′ coupling constants in 13C, 15N-labeled proteins

Summary Sensitive three-dimensional NMR experiments, based on the E.COSY principle, are presented for the measurement of the ³J(H^N,H) and ³J(H^N,C) coupling constants in uniformly ¹³C- and ¹⁵N-labeled proteins. They employ gradient coherence selection in combination with the sensitivity enhancement method in HSQC-type spectra (Cavanagh et al., 1991; Palmer et al., 1991). In most cases, the two measured coupling constants unambiguously define the -angle for protein structure determination. The method is applied to uniformly ¹³C, ¹⁵N-labeled ribonuclease T₁.     

Measurement of Three-bond, 13C′-13Cβ J Couplings in Human Ubiquitin by a Triple Resonance, E. COSY-type NMR Technique

A [CO]HN(CA)CB-E.COSY pulse scheme is described for measurement of three-bond couplings, 3JCC, between carbonyl and aliphatic C carbons in ubiquitin, uniformly enriched with 13C and 15N. A Karplus relation, 3JCC = 1.28 cos2( - 120°) -1.02 cos( - 120°) +0.30 Hz, is obtained by correlating the 3JCC values measured for human ubiquitin with backbone angles from its crystal structure. As predicted, the new Karplus parametrization yields 3JCC values slightly larger than previously obtained by quantitative J correlation [Hu, J.-S. and Bax, A. (1997) J. Am. Chem. Soc., 119, 6360-6368], but considerably smaller than what has been reported on the basis of other E.COSY-type measurements carried out on flavodoxin.     

Accurate measurements of the effects of deuteration at backbone amide positions on the chemical shifts of 15N, 13Cα, 13Cβ, 13CO and 1Hα nuclei in proteins

An approach towards accurate NMR measurements of deuterium isotope effects on the chemical shifts of all backbone nuclei in proteins (¹⁵N, ¹³C_α, ¹³CO, ¹H_α) and ¹³C_β nuclei arising from ¹H-to-D substitutions at amide nitrogen positions is described. Isolation of molecular species with a defined protonation/deuteration pattern at successive backbone nitrogen positions in the polypeptide chain allows quantifying all deuterium isotope shifts of these nuclei from the first to the fourth order. Some of the deuterium isotope shifts measured in the proteins ubiquitin and GB1 can be interpreted in terms of backbone geometry via empirical relationships describing their dependence on (φ; ψ) backbone dihedral angles. Because of their relatively large variability and notable dependence on the protein secondary structure, the two- and three-bond ¹³C_α isotope shifts, ²ΔC_α(N_iD) and ³ΔC_α(N_i+1D), and three-bond ¹³C_β isotope shifts, ³ΔC_β(N_iD), are useful reporters of the local geometry of the protein backbone.     

1H, 13C, and 15N resonance assignments of human ζ-COP

The ζ-COP is one subunit of the COP I coatomer, which mediates the protein trafficking from the cis-Golgi complex to the endoplasmic reticulum and also functions in the intra-Golgi trafficking. The NMR assignments of the ζ-COP are essential for its solution structure determination.     

Determination of 13Cα relaxation times in uniformly 13C/15N-enriched proteins

Summary Relaxation times of ¹³C carbons of uniformly ¹³C/¹⁵N-enriched probes have been investigated. The relaxation behaviour was analyzed in terms of a multispin system. Pulse sequences for the determination of T₁, T₂ and the heteronuclear NOE of ¹³C in uniformly ¹³C/¹⁵N-enriched ribonuclease T1 are presented. The experiments performed in order to obtain T₁ and the heteronuclear NOE were similar to those of the corresponding ¹⁵N experiments published previously. The determination of T₂ for the C-carbon in a completely labeled protein is more complicated, since the magnetization transfer during the T₂ evolution period owing to the scalar coupling of C–C must be suppressed. Various different pulse sequences for the T₂ evolution period were simulated in order to optimize the bandwidth for which reliable T₂ relaxation times can be obtained. A proof for the quality of these pulse sequences is given by fitting the intensity decay of individual ¹H–¹³C cross peaks, in a series of (¹H, ¹³C)-ct-HSQC spectra with a modified CPMG sequence as well as a T_1p sequence for the transverse relaxation time, to a single exponential using a simplex algorithm.     

Identification of Zinc-ligated Cysteine Residues Based on 13Cα and 13Cβ Chemical Shift Data

Although a significant number of proteins include bound metals as part of their structure, the identification of amino acid residues coordinated to non-paramagnetic metals by NMR remains a challenge. Metal ligands can stabilize the native structure and/or play critical catalytic roles in the underlying biochemistry. An atom’s chemical shift is exquisitely sensitive to its electronic environment. Chemical shift data can provide valuable insights into structural features, including metal ligation. In this study, we demonstrate that overlapped ¹³Cβ chemical shift distributions of Zn-ligated and non-metal-ligated cysteine residues are largely resolved by the inclusion of the corresponding ¹³Cα chemical shift information, together with secondary structural information. We demonstrate this with a bivariate distribution plot, and statistically with a multivariate analysis of variance (MANOVA) and hierarchical logistic regression analysis. Using 287 ¹³Cα/¹³Cβ shift pairs from 79 proteins with known three-dimensional structures, including 86 ¹³Cα and¹³Cβ shifts for 43 Zn-ligated cysteine residues, along with corresponding oxidation state and secondary structure information, we have built a logistic regression model that distinguishes between oxidized cystines, reduced (non-metal ligated) cysteines, and Zn-ligated cysteines. Classifying cysteines/cystines with a statisical model incorporating all three phenomena resulted in a predictor of Zn ligation with a recall, precision and F-measure of 83.7%, and an accuracy of 95.1%. This model was applied in the analysis of Bacillus subtilis IscU, a protein involved in iron–sulfur cluster assembly. The model predicts that all three cysteines of IscU are metal ligands. We confirmed these results by (i) examining the effect of metal chelation on the NMR spectrum of IscU, and (ii) inductively coupled plasma mass spectrometry analysis. To gain further insight into the frequency of occurrence of non-cysteine Zn ligands, we analyzed the Protein Data Bank and found that 78% of the Zn ligands are histidine and cysteine (with nearly identical frequencies), and 18% are acidic residues aspartate and glutamate. Electronic Supplementary Material Supplementary material is available for this article at and is accessible for authorized users.     

An HNCO-based Pulse Scheme for the Measurement of 13Cα-1Hα One-bond Dipolar couplings in 15N, 13C Labeled Proteins

A triple resonance pulse scheme is presented for recording 13C-1H one-bond dipolar couplings in 15N, 13C labeled proteins. HNCO correlation maps are generated where the carbonyl chemical shift is modulated by the 13C-1H coupling, with the two doublet components separated into individual data sets. The experiment makes use of recently described methodology whereby the protein of interest is dissolved in a dilute solution of bicelles which orient above a critical temperature, thus permitting measurement of significant couplings (Tjandra and Bax, 1997a). An application to the protein ubiquitin is described.     

13C–13C NOESY spectra of a 480 kDa protein: solution NMR of ferritin

Molecular size has limited solution NMR analyses of proteins. We report ¹³C–¹³C NOESY experiments on a 480 kDa protein, the multi-subunit ferritin nanocage with gated pores. By exploiting ¹³C-resonance-specific chemical shifts and spin diffusion effects, we identified 75% of the amino acids, with intraresidue C–C connectivities between nuclei separated by 1–4 bonds. These results show the potential of ¹³C–¹³C NOESY for solution studies of molecular assemblies >100 kDa.     

1H, 13C, and 15N resonance assignments for human regenerating family Iα protein

Human regenerating (Reg) genes belong to the C-type lectin superfamily and express secretory proteins in various tissues. Reg Iα, also named lithostathine, has multiple roles in numerous biological events such as cytokines, anti-apoptotic factors and the calcium carbonate crystals inhibitor. Under physiological pH, Reg Iα becomes largely insoluble after a self-proteolysis process, and the N-terminally truncated form readily polymerizes into fibrils, which leads to neurodegenerative diseases. Reg Iα may form protofibril via lateral hydrophobic interactions with a native-like conformation. The structural basis from the native to fibril form, as well as the carbohydrate binding sites on Reg Iα, remain unknown. Here we present the NMR backbone and side-chain assignments of Reg Iα for use in further NMR investigations.