The difficulty to automate data acquisition and analysis of complex protein spectra has been one of the major bottlenecks for the widespread use of NMR spectroscopy in structural biology. A promising approach are spectra of high dimensionality (>3) which yield multiple nuclear correlations within fewer experiments, provide high resolution and unambigouos sequential resonance assignment, thus are prone to automation.
Multidimensional spectroscopy (5D-7D) has been explored in solution NMR, however, the concept suffers from a severe inherent contradiction: a satisfactory performance of multiple coherence-transfer experiments is only observed for globular proteins with molecular sizes smaller than about 20 kDa (fast tumbling) or by intrinsically disordered proteins. The deadlock is nowadays removed in proton-detected solid-state NMR at fast magic-angle spinning (MAS). Efficient multiple coherence transfers, narrow proton signals and high detection sensitivity, can be obtained, independently from molecular mass, employing high magnetic fields and ultrafast MAS. The application scope of high-dimensional spectroscopy is thus radically increased.
Here we employ Automated Projection SpectroscopY (APSY), which allows direct inference of a high-dimensional peak list from a number of lower order projection spectra (2D or 3D). We demonstrate the approach with two complementary 5D HN-detected experiments that evolve all traversed backbone nuclei: (H)NCOCANH and (H)NCACONH. We show that sensitive five-dimensional correlations are feasible on microcrystalline and fibrillar proteins at 60 and 110 kHz MAS. APSY, now embedded natively in Bruker TopSpin, not only handles data collection but also entirely bypasses spectral analysis. It delivers an output that directly contains the positions of all resonances. It is coupled to a flexible resonance assignment algorithm FLYA, yielding effortlessly expeditious resonance assignments. The protocol, automated from data collection up to resonance assignment, is in principle amenable to widespread access even by inexperienced spectroscopists, and may push forward the size limits of the proteins amenable to site-specific NMR studies.