Long-observation-window band-selective homonuclear decoupling: Increased sensitivity and resolution in solid-state NMR spectroscopy of proteins

Abstract

Sensitivity and resolution are the two fundamental obstacles to extending solid-state nuclear magnetic resonance to even larger protein systems. Here, a novel long-observation-window band-selective homonuclear decoupling (LOW BASHD) scheme is introduced that increases resolution up to a factor of 3 and sensitivity up to 1.8 by decoupling backbone alpha-carbon (Cα) and carbonyl (C′) nuclei in U-13C-labeled proteins during direct 13C acquisition. This approach introduces short (<200 μs) pulse breaks into much longer (~8 ms) sampling windows to efficiently refocus the J-coupling interaction during detection while avoiding the deleterious effects on sensitivity inherent in rapid stroboscopic band-selective homonuclear decoupling techniques. A significant advantage of LOW BASHD detection is that it can be directly incorporated into existing correlation methods, as illustrated here for 2D CACO, NCO, and NCA correlation spectroscopy applied to the β1 immunoglobulin binding domain of protein G and 3D CBCACO correlation spectroscopy applied to the α-subunit of tryptophan synthase.

13Cα decoupling during direct observation of carbonyl resonances in solution NMR of isotopically enriched proteins

“…Here, we demonstrate that the so-called long-observation-window band-selective homonuclear decoupling method, or LOW-BASHD, recently introduced for solid-state NMR experiments (Struppe et al 2013), can be readily adapted for solution measurements. LOW-BASHD offers advantages over IPAP and SSE in terms of sensitivity and a two-fold reduction in the minimum number of phase cycling steps needed.…”

“…The implementation of LOW-BASHD on a modern NMR spectrometer has been described previously (Struppe et al 2013). Effectively, a series of 180° 13 C α pulses, spaced by intervals 2T, is applied during detection of the 13 C′ FID (Fig.…”

“…Using a digital receiver unit, with rapid oversampling of the FID, the receiver unit is blanked for the duration of the 13 C α pulse plus a short (5 μs) recovery period, which generally amounts to a fraction of the dwell time (reciprocal of the 13 C′ spectral window) used and causes attenuation of the corresponding datapoint. Although in principle the attenuated data-points can be rescaled or otherwise corrected to reduce resulting sidebands in the Fourier transformed spectrum (Struppe et al 2013), we find that this generally is not necessary unless dealing with spectra that have a very high dynamic range.…”

“…We find that decoupling pulses corresponding to (prior to cosinusoidal amplitude modulation) the center lobe of a sinc function are suitable for this purpose, with the duration of the sinc-shaped pulse set to 22600/ν C , where ν C is the 13 C frequency in Hz. The pulse shape and its duration were selected by trial and error, after evaluating a range of different selective pulses and pulse combinations, prior to becoming aware of the earlier solid-state LOW-BASHD work (Struppe et al 2013). The fact that our parameters independently resulted in a shape and duration rather similar to those used in the solid state study suggests that no large further improvement is attainable for the parameters of this pulse scheme.…”

“…When observing 13 C′ at a resolution level of ~25 Hz, as can be reached in advanced solid state NMR studies of polycrystalline proteins, LOW-BASHD decoupling of 13 C α is very robust and relatively insensitive to parameter settings (Struppe et al 2013). However, when aiming for much higher resolution in liquid state NMR, the requirements on decoupling performance become correspondingly more demanding.…”