(13)C-(13)C distance measurements in U-(13)C, (15)N-labeled peptides using rotational resonance width experiment with a homogeneously broadened matching condition

Theory of frequency-selective homonuclear dipolar recoupling in solid-state NMR

In solid-state nuclear magnetic resonance, frequency-selective homonuclear dipolar recoupling is key to quantitative distance measurement or selective enhancement of correlations between atoms of interest in multiple-spin systems, which are not amenable to band-selective or broadband recoupling. Previous frequency-selective recoupling is mostly based on the so-called rotational resonance (R²) condition that restricts the application to spin pairs with resonance frequencies differing in integral multiples of the magic-angle spinning (MAS) frequency. Recently, we have proposed a series of frequency-selective homonuclear recoupling sequences called SPR (short for Selective Phase-optimized Recoupling), which have been successfully applied for selective ¹H-¹H or ¹³C-¹³C recoupling under from moderate (∼10 kHz) to ultra-fast (150 kHz) MAS frequencies. In this study, we fully analyze the average Hamiltonian theory of SPR sequences and reveal the origin of frequency selectivity in recoupling. The theoretical description, as well as numerical simulations and experiments, demonstrates that the frequency selectivity can be easily controlled by the flip angle (p) in the (p)_ϕk(p)_ϕk+π unit in the pSPR-N_n sequences. Small flip angles lead to frequency-selective recoupling, while large flip angles may lead to broadband recoupling in principle. The result shall shed new light on the design of homonuclear recoupling sequences with arbitrary frequency bandwidths.

Can proton-proton recoupling in fully protonated solids provide quantitative, selective and efficient polarization transfer?

Dipolar recoupling sequences have been used to probe spatial proximity of nuclear spins and were traditionally designed to probe rare spins such as ¹³C and/or ¹⁵N nuclei. The multi-spin dipolar-coupling network of the rare spins is weak due to smaller couplings and large chemical shift dispersion. Therefore, the recoupling approaches were tailored to design offset compensated or broadband sequences. In contrast, protons have a substantially stronger dipolar-coupling network and much narrower chemical shift range. Broadband recoupling sequences such as radio-frequency driven recoupling (RFDR), back-to-back (BABA), and lab frame proton-proton spin diffusion have been routinely used to characterize the structures of protein/macromolecules and small molecules. Recently selective ¹H-¹H recoupling sequences have been proposed that combine chemical shift offset of the resolved proton spectrum (at fast MAS) with first- and second-order dipolar recoupling Hamiltonians to obtain quantitative and qualitative proton distances, respectively. Herein, we evaluate the performances of broadband and selective proton recoupling sequences such as finite pulse RFDR (fp-RFDR), band-selective spectral spin diffusion (BASS-SD), second-order cross-polarization (SOCP), and selective recoupling of proton (SERP) in terms of the selectivity and efficiency of ¹H-¹H polarization transfers in a dense network of proton spins and explore the possibility of measuring ¹H-¹H distances. We use theoretical considerations, numerical simulations, and experiments to support the distinct advantages and disadvantages of each recoupling sequence. Experiments were performed on L-histidine.HCl.H₂O at a MAS frequency of 71.43 kHz. This study rationalizes the proper selection of ¹H-¹H recoupling sequences when working with fully protonated solids.

Spinning-rate encoded chemical shift correlations from rotational resonance solid-state NMR experiments

Structural measurements in magic-angle-spinning (MAS) solid-state NMR rely heavily on (13)C-(13)C distance measurements. Broadbanded recoupling methods are used to generate many cross-peaks, but have complex polarization transfer mechanisms that limit the precision of distance constraints and can suffer from weak intensities for distant peaks due to relaxation, the broad distribution of polarization, as well as dipolar truncation. Frequency-selective methods that feature narrow-banded recoupling can reduce these effects. Indeed, rotational resonance (R(2)) experiments have found application in many different biological systems, where they have afforded improved precision and accuracy. Unfortunately, a highly selective transfer mechanism also leads to few cross-peaks in the resulting spectra, which complicates the extraction of multiple constraints. R(2)-width (R(2)W) measurements that scan a range of MAS rates to probe the R(2) matching conditions of one or more sites can improve precision, and also permit multiple simultaneous distance measurements. However, multidimensional R(2)W can be very time-consuming. Here, we present an approach that facilitates the acquisition of 2D-like spectra based on a series of 1D R(2)W experiments, by taking advantage of the chemical shift information encoded in the MAS rates where matching occurs. This yields a more time-efficient experiment with many of the benefits of more conventional multidimensional R(2)W measurements. The obtained spectra reveal long-distance (13)C-(13)C cross-peaks resulting from R(2)-mediated polarization transfer. This experiment also enables the efficient setup and targeted implementation of traditional R(2) or R(2)W experiments. Analogous applications may extend to other variable-MAS and frequency-selective solid-state NMR experiments.

Intramolecular 1H-13C distance measurement in uniformly 13C, 15N labeled peptides by solid-state NMR

A (1)H-(13)C frequency-selective REDOR (FS-REDOR) experiment is developed for measuring intramolecular (1)H-(13)C distances in uniformly (13)C, (15)N-labeled molecules. Theory and simulations show that the experiment removes the interfering homonuclear (1)H-(1)H, (13)C-(13)C and heteronuclear (1)H-(15)N, (13)C-(15)N dipolar interactions while retaining the desired heteronuclear (1)H-(13)C dipolar interaction. Our results indicate that this technique, combined with the numerical fitting, can be used to measure a (1)H-(13)C distance up to 5Å. We also demonstrate that the measured intramolecular (1)H-(13)C distances are useful to determine dihedral angles in proteins.

Homonuclear dipolar recoupling techniques for structure determination in uniformly 13C-labeled proteins

In solid-state NMR magic angle spinning is often used to remove line broadening associated with anisotropic interactions, such as chemical shift anisotropy and dipolar couplings. Dipolar recoupling refers to sequences of pulses designed to reintroduce dipolar interactions that are otherwise averaged by magic angle spinning. One of the key applications of homonuclear (and heteronuclear) dipolar recoupling is for the purpose of protein structure determination. Recoupling experiments, originally designed for applications in spin-pair labeled samples, have been revised in recent years for applications in samples with extensive or uniform incorporation of isotopic labels. In these samples multiple internuclear distances can in principle be probed simultaneously, but the dipolar truncation effects (i.e. attenuation of the effects of weak couplings by strong ones) circumvent such measurements. In this article we review some of the recent developments in homonuclear recoupling methods that allow overcoming this problem.

Targeted 13C-13C distance measurements in a microcrystalline protein via J-decoupled rotational resonance width measurements

Rotational resonance width (R(2)W) magic-angle spinning (MAS) NMR experiments are performed to measure (13)C-(13)C distances in the hydrophobic core of the microcrystalline model protein G(Beta1). Such inter-residue distances are of particular value in NMR structure determinations. The experiments are done at a Larmor frequency of 750 MHz (1)H where the contribution of (13)C chemical shift anisotropy (CSA) to the R(2) transfer mechanism is significant. To minimize line broadening in the 2D spectra, we employ a combination of even/odd isotopic labeling with [1,3-(13)C] glycerol, and J-decoupling in the indirect dimension. This results in high-precision distance measurements between aromatic side chains of three tyrosine residues and distant methyl groups in the hydrophobic core of the protein. Even in the absence of information on the relative orientation of the shift tensors, we obtain relatively high precision data, which can be further improved by additional constraints on the tensor orientations.