De†novo 3D structure determination from sub-milligram protein samples by solid-state 100†kHz MAS NMR spectroscopy.

Solid-state NMR spectroscopy is an emerging tool for structural studies of crystalline, membrane-associated, sedimented, and fibrillar proteins. A major limitation for many studies is still the large amount of sample needed for the experiments, typically several isotopically labeled samples of 10-20 mg each. Here we show that a new NMR probe, pushing magic-angle sample rotation to frequencies around 100 kHz, makes it possible to narrow the proton resonance lines sufficiently to provide the necessary sensitivity and spectral resolution for efficient and sensitive proton detection. Using restraints from such spectra, a well-defined de novo structure of the model protein ubiquitin was obtained from two samples of roughly 500 µg protein each. This proof of principle opens new avenues for structural studies of proteins available in microgram, or tens of nanomoles, quantities that are, for example, typically achieved for eukaryotic membrane proteins by in-cell or cell-free expression.

Automated solid-state NMR resonance assignment of protein microcrystals and amyloids.

Solid-state NMR is an emerging structure determination technique for crystalline and non-crystalline protein assemblies, e.g., amyloids. Resonance assignment constitutes the first and often very time-consuming step to a structure. We present ssFLYA, a generally applicable algorithm for automatic assignment of protein solid-state NMR spectra. Application to microcrystals of ubiquitin and the Ure2 prion C-terminal domain, as well as amyloids of HET-s(218-289) and α-synuclein yielded 88-97 % correctness for the backbone and side-chain assignments that are classified as self-consistent by the algorithm, and 77-90 % correctness if also assignments classified as tentative by the algorithm are included.

Optimal degree of protonation for ¹H detection of aliphatic sites in randomly deuterated proteins as a function of the MAS frequency.

The (1)H dipolar network, which is the major obstacle for applying proton detection in the solid-state, can be reduced by deuteration, employing the RAP (Reduced Adjoining Protonation) labeling scheme, which yields random protonation at non-exchangeable sites. We present here a systematic study on the optimal degree of random sidechain protonation in RAP samples as a function of the MAS (magic angle spinning) frequency. In particular, we compare (1)H sensitivity and linewidth of a microcrystalline protein, the SH3 domain of chicken α-spectrin, for samples, prepared with 5-25 % H(2)O in the E. coli growth medium, in the MAS frequency range of 20-60 kHz. At an external field of 19.96 T (850 MHz), we find that using a proton concentration between 15 and 25 % in the M9 medium yields the best compromise in terms of sensitivity and resolution, with an achievable average (1)H linewidth on the order of 40-50 Hz. Comparing sensitivities at a MAS frequency of 60 versus 20 kHz, a gain in sensitivity by a factor of 4-4.5 is observed in INEPT-based (1)H detected 1D (1)H,(13)C correlation experiments. In total, we find that spectra recorded with a 1.3 mm rotor at 60 kHz have almost the same sensitivity as spectra recorded with a fully packed 3.2 mm rotor at 20 kHz, even though ~20× less material is employed. The improved sensitivity is attributed to (1)H line narrowing due to fast MAS and to the increased efficiency of the 1.3 mm coil.

Simultaneous acquisition of PAR and PAIN spectra.

We present a scheme that allows the simultaneous detection of PAR and PAIN correlation spectra in a single two-dimensional experiment. For both spectra, we obtain almost the same signal-to-noise ratio as if a PAR or PAIN spectrum is recorded separately, which in turn implies that one of the spectra may be considered additional information for free. The experiment is based on the observation that in a PAIN experiment, the PAR condition is always also fulfilled. The performance is demonstrated experimentally using uniformly (13)C,(15)N-labeled samples of N-f-MLF-OH and ubiquitin.