Finite-pulse radio frequency driven recoupling with phase cycling for 2D 1H/1H correlation at ultrafast MAS frequencies

https://doi.org/10.1016/j.jmr.2014.03.004Get rights and content

Highlights

  • 2D 1H/1H RFDR experiments are demonstrated under ultrafast MAS conditions.

  • Performances of various XY-based phase cycling schemes are compared.

  • Results show that XY4 and its super-cycle XY414 render excellent performances.

  • The use of RFDR-XY414 shortens the recycle delay for faster data acquisition.

Abstract

The first-order recoupling sequence radio frequency driven dipolar recoupling (RFDR) is commonly used in single-quantum/single-quantum homonuclear correlation 2D experiments under magic angle spinning (MAS) to determine homonuclear proximities. From previously reported analysis of the use of XY-based super-cycling schemes to enhance the efficiency of the finite-pulse-RFDR (fp-RFDR) pulse sequence, XY814 phase cycling was found to provide the optimum performance for 2D correlation experiments on low-γ nuclei. In this study, we analyze the efficiency of different phase cycling schemes for proton-based fp-RFDR experiments. We demonstrate the advantages of using a short phase cycle, XY4, and its super-cycle XY414 that only recouples the zero-quantum homonuclear dipolar coupling, for the fp-RFDR sequence in 2D 1H/1H correlation experiments at ultrafast MAS frequencies. The dipolar recoupling efficiencies of XY4, XY414 and XY814 phase cycling schemes are compared based on results obtained from 2D 1H/1H correlation experiments, utilizing the fp-RFDR pulse sequence, on powder samples of U–13C,15N-l-alanine, N-acetyl–15N-l-valyl–15N-l-leucine, and glycine. Experimental results and spin dynamics simulations show that XY414 performs the best when a high RF power is used for the 180° pulse, whereas XY4 renders the best performance when a low RF power is used. The effects of RF field inhomogeneity and chemical shift offsets are also examined. Overall, our results suggest that a combination of fp-RFDR-XY414 employed in the recycle delay with a large RF-field to decrease the recycle delay, and fp-RFDR-XY4 in the mixing period with a moderate RF-field, is a robust and efficient method for 2D single-quantum/single-quantum 1H/1H correlation experiments at ultrafast MAS frequencies.

Introduction

Atomic-resolution structural and dynamics information obtained using solid-state NMR experiments can provide piercing insights into the functional properties of a variety of biological molecules including membrane proteins, amyloid proteins and microcrystalline globular proteins [1], [2], [3], [4], [5], [6], [7], [8]. In studies using magic angle spinning (MAS), recoupling RF pulse sequences are typically used to obtain homonuclear chemical shift correlation spectra of low-γ nuclei (mostly 13C) – enabling the assignment of resonances and interatomic distance measurement – from uniformly or non-selectively labeled proteins. One such recoupling sequence is the finite-pulse radio frequency-driven dipolar recoupling (fp-RFDR), which uses a single 180° pulse at the center of each rotor period [9], [10]. While the simple form of this sequence is easy to implement and attractive, its efficiency is lowered by the pulse imperfections such as resonance offset and RF field inhomogeneity, and the interferences from chemical shift anisotropy (CSA) and heteronuclear dipolar couplings [9], [11]. Previous studies have demonstrated the effectiveness of XY-based phase cycling to overcome the limitations of fp-RFDR. Among various phase cycling schemes reported in the literature (Fig. 1), a recent study reported the best performance of fp-RFDR using XY814 phase cycling [12]. The idea behind XY-based phase cycling for the fp-RFDR pulse sequence can be explained by the symmetry based sequence [11], [13] and its super-cycle. Especially the XY814 phase cycling utilizes global super-cycling to suppress the second-order cross terms. While this phase cycling is well suited for studies on low-γ nuclei, it is not certain if this phase cycling would render similar performance for fp-RFDR experiments on protons. Its long cycle duration may not be desirable and therefore it is worthwhile to examine the efficiencies of different phase cycling schemes for proton-based fp-RFDR experiments.

1H based fp-RFDR experiments on solids are challenging, because of low spectral resolution due to unaveraged residual 1H–1H dipolar coupling interactions and relatively small span of chemical shift. However, recent studies have shown the benefits of using 2D 1H/1H fp-RFDR experiments on mobile solids, such as model membranes, to recover motionally averaged dipolar couplings among protons [14], [15]. In addition, the ability of fast MAS enabled the use of fp-RFDR sequence with high 1H spectral resolution [1], [3]. In this study, we demonstrate the use of 2D 1H/1H fp-RFDR experiments under ultrafast MAS condition and also report an examination on the efficiencies of different phase cycling schemes for the fp-RFDR sequence. The criteria for an optimum phase cycling in the fp-RFDR sequence is two-fold: (a) it should reduce the loss of total longitudinal magnetization during the RFDR mixing time enabling a better signal-to-noise ratio; (b) an efficient phase cycle should provide the maximum cross peak intensity in a 2D 1H/1H correlation spectrum. The cross peak intensity depends on the rate of magnetization exchange via the recoupled 1H–1H dipolar couplings, which is determined by the following two factors. One is the scaling factor of the recoupled 1H–1H dipolar interactions. In the fp-RFDR sequence, this is solely determined by the ratio of the 180° pulse length to the cycle time of sample spinning. The second factor is the loss of magnetization due to relaxation including the RF-field inhomogeneity, chemical shift offset, long-range dipolar couplings, RF-induced sample heating. These factors – particularly the second one – greatly affect the efficiency of the phase cycling used in fp-RFDR experiments.

In this study, we systematically evaluate the performances of various phase cycling schemes for the fp-RFDR sequence. Our results show that a short phase cycle, XY4 corresponding to the R44−1 symmetry [16], and its super-cycle XY414 are well suited for proton-based fp-RFDR experiments. The fp-RFDR using these phase cycling schemes renders zero-quantum homonuclear dipolar recoupling by suppressing all other interactions including CSA, heteronuclear dipolar and scalar couplings, and isotropic chemical shifts. Since the magnetization transfer during the mixing time of a first-order dipolar recoupling sequence is independent of the MAS frequency, the fp-RFDR based 2D chemical shift correlation experiments utilizing these phase cycling schemes have unique advantages in studying biological solids at very fast and ultra-fast MAS frequencies.

Section snippets

Experimental

All NMR experiments were performed on a 600 MHz JNM-ECA600II NMR spectrometer (JEOL RESONANCE Inc., Tokyo, Japan) equipped using a 0.75 mm ultrafast MAS probe. Powder samples of U–13C,15N-l-alanine and glycine were purchased from Isotec and were used without any modification. N-acetyl-15N-l-valyl–15N-l-leucine (NAVL) was prepared as explained elsewhere [17], [18]. Samples were packed in a 0.75 mm zirconia rotor and all experiments were performed at room temperature (ca. 24 °C). Experiments were

Simulations

The spin dynamics simulations were performed using the SPINEVOLUTION software [19]. Three proton spins were considered in the simulations. Distance between any two protons was set at 0.16 nm. The simulations were performed at 80 and 92 kHz spinning speeds with a length of 2 μs for the 180° pulse. As results obtained with the two spinning speeds are similar, only data obtained from 92 kHz MAS are presented. Through simulations, we obtained the build-up of the magnetization transfer efficiency from

Results and discussion

In this study, we systematically investigate the efficiencies of different phase cycling schemes by employing them in the proton-based fp-RFDR pulse sequence (Fig. 1) under ultrafast MAS conditions. Experimental results obtained from powder samples of U–13C,15N-l-alanine and NAVL are given, while those from glycine are not included as they support the results obtained from other compounds and do not provide additional new information. The simulations were performed at 80 and 92 kHz spinning

Conclusions

In this study, we have reported an evaluation of different XY based phase cycling schemes for the fp-RFDR 1H/1H homonuclear recoupling sequence based on ultra-fast (at 80, 92 and 100 kHz) MAS experiments on powder samples of U–13C,15N-l-alanine, glycine and NAVL. We have demonstrated that the efficiency of a phase cycle for a proton-based fp-RFDR experiment is different from that for low-γ (for example 13C) nuclei. For proton-based RFDR experiments, an efficient phase cycle should avoid the loss

Acknowledgments

This research was supported by funds from NIH (GM084018 and GM095640 to A.R.) and JEOL Resonance Inc. (Tokyo, Japan). We would like to thank the JEOL Resonance scientists for help with the spectrometer and ultrafast MAS probe.

References (20)

There are more references available in the full text version of this article.

Cited by (53)

View all citing articles on Scopus
View full text