Slice-selective NMR spectroscopy

NMR spectroscopy is probably the most frequently used technique for the structural investigation of small to medium sized organic and biological molecules. Due to its high sensitivity and widespread occurrence the most frequently measured nucleus is 1H (proton). In contrast to its favorable sensitivity, the resolution of proton NMR spectra is rather poor. This results in part from the limited proton chemical shift range (~10 ppm) but also from extensive signal splitting due to scalar coupling to nearby protons. Pure-shift NMR spectroscopy (by broadband homonuclear decoupling) has been described as an approach to significantly enhance the resolution of proton NMR spectra. This approach leads to proton spectra consisting of single lines, reminiscent of proton-decoupled C-13 spectra. The resulting resolution is comparable to regular 1H spectra acquired at several GHz. We have shown previously that homonuclear broadband decoupled NMR spectra can be obtained by a combination of frequency and spatially-selective pulses. This method, which is also called Zangger- Sterk method, can also be applied during the acquisition of the FID, i.e. in real-time. However, the resulting spectra are much noisier than regular 1D 1H spectra and of course all scalar coupling information is completely lost. Within this project we will establish several techniques aimed at reducing the artifacts in real-time homonuclear broadband decoupled spectra; extracting scalar coupling information with enhanced resolution from partially coupled spectra and speeding up data acquisition. The vast majority of the proposed methods relies on the use of spatially- and frequency- selective (slice-selective) pulses. By this excitation, different signals are excited in individual slices of the NMR sample tube, which allows a homonuclear system to be transformed to a spatially-selective pseudo-heteronuclear ensemble. This enables the application of heteronuclear NMR methods to homonuclear spin systems.

NMR spectroscopy is one of the most useful analytical techniques for the structural characterization of organic and biomolecules. Due to its widespread occurrence and high sensitivity, hydrogen nuclei (protons) are most often used in NMR spectroscopy. However, the resolution of proton NMR is rather limited. The main reason for that are signal splitting by scalar coupling between neighboring hydrogen nuclei. The aim of pure shift NMR is the removal of all signal splittings to enhance the resolution of hydrogen NMR spectra. One of the more often used pure shift techniques is the spatially-selective decoupling (called Zangger-Sterk method in the literature), which was developed in our lab. Within this project we could significantly reduce the artifacts which are typical for these kind of spectra and enhance the sensitivity, which is significantly below conventional NMR spectra. Scalar coupling, which is completely removed in pure shift spectra does contain important structural information. Therefore, we developed a technique which does not remove but downscale couplings. Sometimes it is more useful to determine scalar couplings with the highest resolution possible. A method which enables just that has been developed during this project- real-time J-upscaling. Thereby, during NMR data acquisition all homonuclear couplings are enlarged by the same desired factor. This results in an actual resolution enhancement of structurally relevant couplings, since other phenomena, like signal broadening by magnetic field inhomogeneities are not enhanced. Couplings which are hidden in the linewidth of regular spectra can thus be determined. Using the methods, which have been improved or newly developed within this project we are providing tools to record NMR spectra with higher resolution with respect to both signal separation (chemical shift) but also signal splittings (scalar coupling). Such spectra can be used not only to acquire more accurate structures of organic molecules and compound mixtures, but should also prove useful for higher quality protein ligand NMR spectra. This in turn should result in higher accuracy ligand structures, which are needed for example for structure guided drug design.