Venkataraman Thanabal

A constant-time three-dimensional triple-resonance pulse scheme to correlate intraresidue 1HN, 15N, and 13C′ chemical shifts in 15N13C-labelled proteins

Sequence-specif ic resonance assignments are a prerequisite for structural and dynamical interpretation of protein NMR spectra. For proteins smaller than 10 kDa assignment strategies have relied upon through-bond correlations in homonuclear COSY and TOCSY spectra to identify resonances associated with particular spin systems. Conformation-dependent nuclear Overhauser effects are then emp loyed to sequentially connect these spin systems (1-5). In larger proteins, however, extensive resonance overlap and decreased sensitivity of experiments utilizing ‘H‘H scalar couplings have hindered this approach. ‘HI%15N triple-resonance experiments provide a conformation-independent approach for the assignment of backbone resonances in I%15N-labeled large proteins (6-13). In addition these experiments allow accurate measurement of coupling constants in proteins with large linewidths (14, 15). These experiments exploit large heteronuclear one-bond couplings to transfer magnetization with the sensitivity of indirect detection. As demonstrated in calmodulin ( 16, 17) backbone assignments utilize four tr iple-resonance experiments [ HNCa, HNCO, HCaCO, and HCA(CO)N] and the 3D TOCSY-HMQC experiment ( 18). A fifth tr iple-resonance experiment [ H( CA)NHN] is a useful complement to the TOCSY-HMQC experiment, correlating ‘HN “N and ‘H” resonances ( 7, 9, 13). Furthermore, a sixth experiment, the HN (CO ) CA. has been introduced, providing additional sequential information ( 11) . Of the experiments listed above, those which detect am ide protons [ HNCA, HNCO, H(CA)NHN, HN(CO)Ca] need to be collected in H20, while those with detection of (Y protons [ HCACO and HCA( CO)N] are best performed in 2H20. This causes difficulties if assignments are based on al ignment of C’ chemical shifts, because these resonances may show significant isotope shifts, depending on whether the carbonyl is hydrogen bonded to a proton or a deuteron. It is therefore desirable to perform the ma jority of experiments in the same solvent ( H20), restricting oneself to those experiments which detect am ide protons [ HNCA, HNCO, HN( CO)CA. H( CA )NHN 1.

A study of Src SH2 domain protein-phosphopeptide binding interactions by electrospray ionization mass spectrometry

The noncovalent binding of various peptide ligands to pp60src (Src) SH2 (Src homology 2) domain protein (12.9 ku) has been used as a model system for development of electrospray ionization mass spectrometry (ESI-MS) as a tool to study noncovalently bound complexes. SH2 motifs in proteins are critical in the signal transduction pathways of the tyrosine kinase growth factor receptors and recognize phosphotyrosine-containing proteins and peptides. ESI-MS with a magnetic sector instrument and array detection has been used to detect the protein-peptide complex with low-picomole sensitivity. The relative abundances of the multiply charged ions for the complex formed between Src SH2 protein and several nonphosphorylated and phosphorylated peptides have been compared. The mass spectrometry data correlate well to the measured binding constants derived from solution-based methods, indicating that the mass spectrometry-based method can be used to assess the affinity of such interactions. Solution-phase equilibrium constants may be determined by measuring the amount of bound and unbound species as a function of concentration for construction of a Scatchard graph. ESI-MS of a solution containing Src SH2 with a mixture of phosphopeptides showed the expected protein-phosphopeptide complex as the dominant species in the mass spectrum, demonstrating the method’s potential for screening mixtures from peptide libraries.

A new 3D HN(CA)HA experiment for obtaining fingerprint HN-Hα cross peaks in15N- and13C-labeled proteins

SummaryA new 3D1H−15N−13C triple resonance experiment is presented that provides in-phase absorptive cross peaks between amide protons and α-protons of the same and the sequentially preceding residue. The experiment yields similar connectivities as those described previously by Montelione and Wagner (1990a) [J. Magn. Reson.,87, 183–188] and Kay et al. (1991) [J. Magn. Reson.,91, 84–92]. However, the pulse sequence was designed to minimize the time that transverse coherence of the13Cα nucleus is present, since this nucleus has the shortest transverse relaxation time of all the nuclei involved in these experiments. This is achieved by using a coherence transfer pathway from1HN to15N,13Cα,1Hα and back to the1HN. In the sequence described, transverse13Cα coherence is present only for a length of ca. I1J(Cα-Hα). This reduces loss of signal due to transverse relaxation. We tested the technique on uniformly15N- and13C-enriched T4 lysozyme.

2D Heteronuclear NMR Measurements of Spin-Lattice Relaxation Times in the Rotating Frame of X Nuclei in Heteronuclear HX Spin Systems

Theoretical and experimental aspects of T,, are discussed for a heteronuclear HX twospin system (T

Improved accuracy of heteronuclear transverse relaxation time measurements in macromolecules. elimination of antiphase contributions

) where only the X nucleus is spin-locked. An expression for TE in terms of spectral density functions and the effective magnetic field parameters is developed. It shows that T:, offers potentially dierent information about the spectral densities than either r:, TF, or the steady-state NOEX. We present a 2D heteronuclear NMR pulse sequence for measuring site-specific T

The 13C chemical shifts of amino acids in aqueous solution containing organic solvents: Application to the secondary structure characterization of peptides in aqueous trifluoroethanol solution

‘s in biomolecules. The sequence is based on a double-INEPT transfer and applies a spin lock to the heteronuclei for variable delays. If a weak on-resonance spin lock is used, and if the spectral density functions are assumed to be Lorentzians, then rc is theoretically indistinguishable from r:. We conclude with an application of the pulse sequence to the uniformly ‘5N-enriched protein eglin c. The rc, data reflect the differential mobility in the molecule. o 199 t Academic PESS, h.

The solution structure of omega-Aga-IVB, a P-type calcium channel antagonist from venom of the funnel web spider, Agelenopsis aperta.

Heteronuclear T2 relaxation measurements ( 13C, “N) play an integral role in the characterization of internal motions in proteins by NMR (l-3). Recently there has been some controversy concerning the correct method of measuring “N or “C T2 values because different experiments produced different (mainly too short) T2 values (1-3). We propose and show experimental evidence that this is due to an oscillation between in-phase and antiphase coherence of the heteronucleus if chemical-shift refocusing and concomitant proton decoupling are achieved via a single ( Carr-Purcell). or a train of ( Meiboom-Gill) X-nucleus 180” pulses. This leads to a mixture of contributions from in-phase and antiphase relaxation, and the latter causes apparently faster T2 relaxation. We propose to eliminate the antiphase contributions from the T2 measurements by applying a spin lock on the heteronucleus. The result is a significant improvement in the accuracy of heteronuclear T2 measurements. A basic 2D pulse sequence for measuring heteronuclear T2 values in a 2D NMR experiment ( 1) is shown in Fig. 1 a. We are mainly concerned with “N relaxation and denote the heteronucleus as N and follow the product-operator formalism (4). In the pulse sequence of Fig. 1 a, in-phase transverse coherence, N,, relaxes during a refocusing delay 7, primarily due to dipole-dipole interactions between the N spins and their directly bonded protons (I spins). “N relaxation studies on the proteinase inhibitor . . . . . . eglin c reveal that the use of different refocusing pulse schemes (Fig. la: 1, u, m) yields different transverse relaxation times (5). Specifically, the use of a single 180” pulse produces, in small proteins, relaxation times about 50% shorter than those obtained using a 3 kHz spin lock. Carr-Purcell-Meiboom-Gill (CPMG) (6, 7) relaxation times yield intermediate values, and the apparent T2 values increase as the spacing between the 180” pulses is narrowed. However, with our spectrometer, we never came closer than to about 75% of the spin-lock values, being limited by a hard-pulse duty cycle of 2%. For all residues in the protein, the maximum transverse relaxation time is

The solution structure of a cyclic endothelin antagonist, BQ-123, based on 1H-1H and 13C-1H three bond coupling constants

SummaryThe 13C chemical shifts for all of the protonated carbons of the 20 common amino acid residues in the protected linear pentapeptide Gly-Gly-X-Gly-Gly have been obtained in water at low pH as well as in aqueous solution containing 10, 20 and 30% acetonitrile or trifluoroethanol. Dioxane was used as an internal reference and its carbon chemical shift value was found to be 66.6 ppm relative to external TMS in water. Comparison of the different referencing methods for 13C chemical shifts in organic cosolvent mixtures showed that an external standard (either TMS or TSP capillary) was the most appropriate. In the present study, external TSP was chosen to define the 0 ppm of the 13C chemical shift scale. When the difference in referencing the dioxane carbon resonance is taken into account, the carbon chemical shift values of the amino acids in aqueous solution are similar to those previously reported (Richarz and Wüthrich (1978) Biopolymers, 17, 2133–2141; Howarth and Lilley (1979) Prog. NMR Spectrosc., 12, 1–40). The pentapeptides studied were assumed to be in a random coil conformation and the measured 13C chemical shifts were used as reference values to correlate carbon chemical shifts with the secondary structure of two well-characterized peptides, bombesin and the 1–29 amino acid fragment of Nle27 human growth hormone-releasing factor. In both cases, the Cα chemical shifts exhibited a characteristic positive deviation from the random coil values, which indicates the presence of α-helices.

A new 1H15N13C triple-resonance experiment for sequential assignments and measuring homonuclear HαHN vicinal coupling constants in polypeptides

SummaryThe 48 amino acid peptides ω-Aga-IVA and ω-Aga-IVB are the first agents known to specifically block P-type calcium channels in mammalian brain, thus complementing the existing suite of pharmacological tools used for characterizing calcium channels. These peptides provide a new set of probes for studies aimed at elucidating the structural basis underlying the subtype specificity of calcium channel antagonists. We used 288 NMR-derived constraints in a protocol combining distance geometry and molecular dynamics employing the program DGII, followed by energy minimization with Discover to derive the three-dimensional structure of ω-Aga-IVB. The toxin consists of a well-defined core region, comprising seven solvent-shielded residues and a well-defined triple-stranded β-sheet. Four loop regions have average backbone rms deviations between 0.38 and 1.31 Å, two of which are well-defined type-II β-turns. Other structural features include disordered C- and N-termini and several conserved basic amino acids that are clustered on one face of the molecule. The reported structure suggests a possible surface for interaction with the channel. This surface contains amino acids that are identical to those of another known P-type calcium channel antagonist, ω-Aga-IVA, and is rich in basic residues that may have a role in binding to the anionic sites in the extracellular regions of the calcium channel.

Refined solution structure of the DNA-binding domain of GAL4 and use of 3J(113Cd,1H) in structure determination

A cyclic pentapeptide endothelin antagonist, cyclo(dTrp‐dAsp‐Pro‐dVal‐Leu), recently reported (K. Ishikawa et al., 13th Am. Pept. Symp., Cambridge MA, 1991) has been studied by NMR spectroscopy and molecular modeling. A stable structure has been determined without the use of nuclear Overhauser effects and is based primarily on homonuclear and heteronuclear three bond coupling constants. The 13C‐edited TOCSY experiment is demonstrated at natural abundance and ∼30 mM peptide concentrations. Three bond 13C–1H coupling constants obtained by this method are shown to reduce the ambiguity in φ angle determination which exists when only interproton coupling constants are used. Three out of four φ angles were determined uniquely by this method and the fourth was reduced to two possible values. The proline φ angle was determined to be −78° based on the 3 J HzHα and 3 J HzHβ coupling constants. Comparison of amide proton temperature dependence, chemical shifts and vicinal proton coupling constants in a 20% acetonitrile/80% water solvent mixture and in (CD3)2SO indicates that the structure is similar in both solvents.