High-field ex vivo and in vivo two-dimensional nuclear magnetic resonance spectroscopy in murine brain: Resolving and exploring the molecular environment.

The structural and chemical complexities within the brain pose a challenge that few noninvasive techniques can tackle with the dexterity of nuclear magnetic resonance (NMR) spectroscopy. Still, even with the advent of ultrahigh fields and of cryogenically cooled coils for in vivo research, the superposition of metabolic resonances arising from the brain remains a challenge. The present study explores the potential to tackle this milieu using a combination of two-dimensional (2D) NMR techniques, implemented on murine brains in vivo at 15.2 T and ex vivo at 14.1 T. While both experiments were affected by substantial inhomogeneous broadenings conveying distinct elongated lineshapes to the cross-peaks, the ability of increased fields to resolve off-diagonal resonances was clear. A comparison between the corresponding conventional and double quantum-filtered correlated spectroscopy traces enabled an improved assignment of in vivo resonances on the basis of more sensitive ex vivo 2D acquisitions, foremost on the basis of homonuclear cross-relaxation-driven correlations for peaks resonating downfield from water, and of heteronuclear correlations at natural abundance for the upfield protons. With the aid of such 2D correlations approximately 29 metabolites could be resolved and identified. This enhanced resolution was used to explore features related to the metabolites' diffusivities, their exposure to water, and their facility to undergo magnetization transfers to amide/amine/hydroxyl resonances. Cross-peaks from main murine brain biomolecules, including choline, creatine, γ-aminobutyric acid, N-acetyl aspartate, glutamine, and glutamate, showed enhancements in several of these various features, opening interesting vistas about metabolite compartmentalization as viewed by these 2D NMR experiments.

Measuring strong one-bond dipolar couplings using REDOR in magic-angle spinning solid-state NMR.

Rotational-Echo DOuble Resonance, REDOR, is an experimentally robust and a well-established dipolar-recoupling technique to measure dipolar couplings between isolated pairs of spin-1/2 heteronuclei in solid-state nuclear magnetic resonance. REDOR can also be used to estimate motional order parameters when the bond distance is known, for example, in the case of directly bound nuclei. However, the relatively fast dipolar dephasing for strongly coupled spin-1/2 pairs, such as 13C-1H, makes the stroboscopic measurement required in this experiment challenging, even at fast Magic-Angle-Spinning (MAS) frequencies. In such cases, modified REDOR-based methods like Shifted-REDOR (S-REDOR) are used to scale the dipolar coupling compared to REDOR. This is achieved by changing the position of one of the two recoupling π-pulses in a rotor period. This feature, however, comes at the cost of mixing multiple Fourier components of the dipolar coupling and can, additionally, require high radio-frequency amplitudes to realise small scaling factors. We introduce here a general pulse scheme which involves shifting both the π pulses in the REDOR scheme to achieve arbitrary scaling factors whilst retaining the robustness and simplicity of REDOR recoupling and avoiding the disadvantages of S-REDOR. The classical REDOR is a specific case of this scheme with a scaling factor of one. We demonstrate the results on isolated 13C-15N and 1H-13C spin pairs at 20 and 62.5 kHz MAS, respectively.

Sine-squared shifted pulses for recoupling interactions in solid-state NMR.

Rotational-Echo DOuble-Resonance (REDOR) is a versatile experiment for measuring internuclear distance between two heteronuclear spins in solid-state NMR. At slow to intermediate magic-angle spinning (MAS) frequencies, the measurement of distances between strongly coupled spins is challenging due to rapid dephasing of magnetisation. This problem can be remedied by employing the pulse-shifted version of REDOR known as Shifted-REDOR (S-REDOR) that scales down the recoupled dipolar coupling. In this study, we propose a new variant of the REDOR sequence where the positions of the π pulses are determined by a sine-squared function. This new variant has scaling properties similar to S-REDOR. We use theory, numerical simulations, and experiments to compare the dipolar recoupling efficiencies and the experimental robustness of the three REDOR schemes. The proposed variant has advantages in terms of radiofrequency field requirements at fast MAS frequencies.

Measuring Dipolar Order Parameters in Nondeuterated Proteins Using Solid-State NMR at the Magic-Angle-Spinning Frequency of 100 kHz.

Proteins are dynamic molecules, relying on conformational changes to carry out function. Measurement of these conformational changes can provide insight into how function is achieved. For proteins in the solid state, this can be done by measuring the decrease in the strength of anisotropic interactions due to motion-induced fluctuations. The measurement of one-bond heteronuclear dipole-dipole coupling at magic-angle-spinning (MAS) frequencies >60 kHz is ideal for this purpose. However, rotational-echo double resonance (REDOR), an otherwise gold-standard technique for the quantitative measurement of these couplings, is difficult to implement under these conditions, especially in nondeuterated samples. We present here a combination of strategies based on REDOR variants ϵ-REDOR and DEDOR (deferred REDOR) and simultaneously measure residue-specific 15N-1H and 13Cα-1Hα dipole-dipole couplings in nondeuterated systems at the MAS frequency of 100 kHz. These strategies open up avenues to access dipolar order parameters in a variety of systems at the increasingly fast MAS frequencies that are now available.

Overcoming Prohibitively Large Radiofrequency Demands for the Measurement of Internuclear Distances with Solid-State NMR under Fast Magic-Angle Spinning.

Solid-state NMR is a powerful tool to measure distances and motional order parameters which are vital tools in characterizing the structure and dynamics of molecules. Magic-angle spinning (MAS), widely employed in solid-state NMR, averages out dipole-dipole couplings that carry such information. Hence, rotor-synchronized radiofrequency (RF) pulses, that interfere with MAS averaging, are commonly employed to measure such couplings. However, most of the methods that achieve this, rotational echo double resonance (REDOR) being a classic example, require RF amplitudes that are greater than or equal to the MAS frequency. While feasible at MAS frequencies <40 kHz, these requirements become prohibitively large for higher MAS frequencies (40-110 kHz), which are now commercially available. Here, we redesign the REDOR experiment so that RF amplitudes as low as 0.5-0.7 times the spinning frequency can be used. This sequence, name deferred rotational echo double resonance (DEDOR), thus extends the utility of this method to the fastest MAS frequencies currently commercially available (111 kHz). The generality of this strategy is shown by extending it to other methods that utilize the same principle as REDOR. They will be useful in obtaining structural parameters for a wide range of molecules using solid-state NMR under fast MAS with the additional advantage of higher spectral resolution under these conditions.

On the direct relation between REDOR and DIPSHIFT experiments in solid-state NMR.

Rotational-echo double resonance (REDOR) and Dipolar-coupling chemical-shift correlation (DIPSHIFT) are commonly used experiments to probe heteronuclear dipole-dipole couplings between isolated pairs of spin-12 nuclei in magic-angle-spinning (MAS) solid-state NMR. Their widespread use is due to their robustness to experimental imperfections and a straightforward interpretation of data. Both of these experiments use rotor-synchronised π pulses to recouple the heteronuclear dipole-dipole couplings, and the observed intensity of resonances is modulated by a recoupled phase factor depending on the position or duration of the recoupling pulses. Several modifications to both of these experiments have been proposed, for example, the development of DIPSHIFT which employs strategies that mimic the multi-rotor-period nature of REDOR. We show here that REDOR and DIPSHIFT are in fact alternate implementations of the same experiment. The overt similarity in the design of REDOR and DIPSHIFT is also reflected in their theoretical description. Dipolar dephasing curves in REDOR are obtained by increasing the recoupling duration whilst keeping the position of the pulses constant, which results in a dephasing factor that is a function of only the dephasing time. DIPSHIFT, on the other hand, is a constant-time version of REDOR; the dipolar dephasing is a function of the position of the pulses with respect to the rotor period. We discuss the advantages and disadvantages of each implementation and suggest domains of applicability for these sequences.

Refocusing pulses: A strategy to improve efficiency of phase-modulated heteronuclear decoupling schemes in MAS solid-state NMR.

The strategy of using π pulses in conjunction with continuous-wave radio-frequency fields to refocus spin interactions has lead to robust and efficient family of heteronuclear decoupling schemes in magic-angle spinning solid-state NMR, denoted as, rCW schemes. Here, we investigate the generality of the application of such refocussing pulses in other phase-modulated decoupling schemes, notably the super-cycled XiX decoupling. XiX is a commonly used heteronuclear decoupling scheme under conditions of fast MAS and low-amplitude radio-frequency irradiation. The refocussing of interactions is achieved by inserting π pulses with a phase of 135° in the supercycled XiX scheme. The refocussed XiX, rXiX, scheme has improved decoupling efficiency, better offset tolerance, and easier experimental setup compared to the XiX scheme.

Selective 1H-1H Distance Restraints in Fully Protonated Proteins by Very Fast Magic-Angle Spinning Solid-State NMR.

Very fast magic-angle spinning (MAS > 80 kHz) NMR combined with high-field magnets has enabled the acquisition of proton-detected spectra in fully protonated solid samples with sufficient resolution and sensitivity. One of the primary challenges in structure determination of protein is observing long-range 1H-1H contacts. Here we use band-selective spin-lock pulses to obtain selective 1H-1H contacts (e.g., HN-HN) on the order of 5-6 Å in fully protonated proteins at 111 kHz MAS. This approach is a major advancement in structural characterization of proteins given that magnetization can be selectively transferred between protons that are 5-6 Å apart despite the presence of other protons at shorter distance. The observed contacts are similar to those previously observed only in perdeuterated proteins with selective protonation. Simulations and experiments show the proposed method has performance that is superior to that of the currently used methods. The method is demonstrated on GB1 and a ß-barrel membrane protein, AlkL.